Skip to content
EA Crux Project

External Resources

Browse all external resources (papers, books, blog posts, reports, etc.) referenced throughout the knowledge base. Use the filters to find resources by type or search by title/author.

3133 resourcesπŸ”— 2393 websπŸ“„ 360 papersπŸ›οΈ 245 governments✏️ 88 blog postsπŸ“– 34 referencesπŸ“‹ 9 reports🎀 4 talks
Showing 3133 of 3133 resources
TypeTitleCredibilityContentAuthorsDateTagsCited ByNotes
πŸ“– Reference"alignment faking"Wikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”alignment1 articleView β†’
πŸ“„ Paper"Are Emergent Abilities a Mirage?"arXivβ˜…β˜…β˜…β˜†β˜†SummaryRylan Schaeffer, Brando Miranda, Sanmi Koyejo2023-04-28capabilitiesllm+33 articlesView β†’
πŸ”— Web"bitter lesson" phenomenonβ€”Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— Web"Governing AI for Humanity"United Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”democratic-innovationcollective-intelligence+12 articlesView β†’
πŸ”— Web"How Governments Use Facial Recognition for Protest Surveillance."β€”Noneβ€”β€”authoritarianismstability+11 articleView β†’
πŸ”— Web"humans, not AI" should control nuclear weaponsBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationtiming+13 articlesView β†’
✏️ Blog Post"Intellectual Debt"Mediumβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ“„ Paper"More Is Different"Scienceβ˜…β˜…β˜…β˜…β˜…MetadataP. Anderson1972-08-04scalingcapability-evaluation+11 articleView β†’
πŸ”— Web"Most Important Century"β€”Noneβ€”β€”effective-altruismai-safety-funding+11 articleView β†’
✏️ Blog Post"On AGI Ruin: A List of Lethalities"β€”Noneβ€”β€”agicapability-generalization+21 articleView β†’
πŸ“„ Paper"Optimal Policies Tend To Seek Power"arXivβ˜…β˜…β˜…β˜†β˜†SummaryChien-Ping Lu2025-01-04β€”1 articleView β†’
πŸ”— Web"Retrospective on My Posts on AI Threat Models"β€”Noneβ€”β€”capability-generalizationalignment-stability+11 articleView β†’
✏️ Blog Post"Situational Awareness"LessWrongβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”intelligence-explosionrecursive-self-improvement+12 articlesView β†’
πŸ”— Web"The case for ensuring that powerful AIs are controlled" (May 2024)β€”Noneβ€”β€”interpretabilityscalable-oversight+12 articlesView β†’
πŸ”— Web"The OpenAI Files" reveals deep leadership concerns about Sam Altman and safety failuresFortune

The 'OpenAI Files' examines internal issues at OpenAI, highlighting leadership challenges and potential risks in AI development. The report critiques Sam Altman's leadership and the company's evolving approach to ethical AI.

β˜…β˜…β˜…β˜†β˜†FullBeatrice Nolanβ€”safety1 articleView β†’
πŸ”— Web"The Wrong Kind of AI"β€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ”— Web"we are past the event horizon; the takeoff has started"β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Web"Will We Run Out of Data?"Epoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-forecastingcompute-trends+11 articleView β†’
πŸ”— Web**EU AI Office**European Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorcompetition+39 articlesView β†’
πŸ”— Web**Future of Humanity Institute**Future of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”talentfield-building+330 articlesView β†’
πŸ”— Web\$10 billion additional investmentCNBCβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”chatgptgpt-4+11 articleView β†’
πŸ”— Web\$10-100M per training runβ€”Noneβ€”β€”training1 articleView β†’
πŸ”— Web\$109 billion in 2024β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Web\$10M in AI centersβ€”Noneβ€”β€”compute-thresholdsgovernance+11 articleView β†’
πŸ”— Web\$124 million Series Aβ€”Noneβ€”β€”chatgptgpt-4+11 articleView β†’
πŸ”— Web\$1M actually availableβ€”Noneβ€”β€”compute-thresholdsgovernance+11 articleView β†’
πŸ”— Web$10+ billion in philanthropic commitmentsOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ”— Web$100 million and 25,000+ GPUsβ€”Noneβ€”β€”computegovernance+32 articlesView β†’
πŸ”— Web$11 million in fundingβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ Government$16.6 billion in 2024β€”Noneβ€”β€”social-engineeringvoice-cloning+11 articleView β†’
πŸ”— Web$2.1 million from Open Philanthropy in 2019Open Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”research-agendasalignment+11 articleView β†’
πŸ”— Web$23.6M in Open Philanthropy fundingOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”field-buildingtraining-programs+11 articleView β†’
πŸ”— Web$265,000 from Open Philanthropy in March 2022Open Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”research-agendasalignment+11 articleView β†’
πŸ”— Web$3 billion from GoogleCNBCβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”ai-safetyconstitutional-ai+11 articleView β†’
πŸ”— Web$580M investment in AnthropicAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”effective-altruismai-safety-funding+11 articleView β†’
πŸ”— Web$61.5 billionCNBCβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”ai-safetyconstitutional-ai+11 articleView β†’
πŸ”— Web$67.2 billion in AI investmentBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic-inequalitymarket-concentration+11 articleView β†’
πŸ”— Web$7.5 million grantβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— Web$70-100 millionβ€”Noneβ€”β€”regulationstate-policy+11 articleView β†’
πŸ”— Web$8 billion from Amazonβ€”Noneβ€”β€”ai-safetyconstitutional-ai+11 articleView β†’
πŸ”— Web€400M budgetEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— Web10 million features extractedAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interpretabilityscalable-oversight+12 articlesView β†’
πŸ”— Web10-15% of market by 2030β€”Noneβ€”β€”tool-useagentic+11 articleView β†’
πŸ”— Web10-42% correct root cause identificationAnthropic Alignmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— Web10-year moratorium on state and local AI lawsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“– Reference113+ current and former employeesWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”regulationstate-policy+11 articleView β†’
πŸ”— Web15+ citationsGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-safetyx-risk+11 articleView β†’
πŸ”— Web16 companies committed to publish frontier AI safety protocolsβ€”Noneβ€”β€”safety1 articleView β†’
πŸ”— Web2.0β€”Noneβ€”β€”ai-safetyconstitutional-ai+11 articleView β†’
πŸ”— Web2.2β€”Noneβ€”β€”ai-safetyconstitutional-ai+12 articlesView β†’
πŸ“– Reference200-500 millisecondsWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ”— Web200+ million Uyghurs under surveillanceβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— Web2019 Iranian GPS spoofing incidentReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— Web2022 reportβ€”Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ”— Web2023 AI Impacts surveyAI Impactsβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— Web2023 AI regulationsβ€”Noneβ€”β€”governancex-risk+21 articleView β†’
πŸ“– Reference2023 AI researcher surveyWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ“„ Paper2023 AI researcher surveyarXivβ˜…β˜…β˜…β˜†β˜†SummaryKatja Grace, Harlan Stewart, Julia Fabienne SandkΓΌhler, Stephen Thomas, Ben Weinstein-Raun, Jan Brauner, Richard C. Korzekwa2024-01-05x-risktraining+21 articleView β†’
πŸ”— Web2023 America in One Room: Democratic Reformβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— Web2023 Expert Survey on AI RiskAI Impactsβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“„ Paper2024 Chemistry Nobel: AlphaFold - NatureNature

Google DeepMind's John Jumper and Demis Hassabis, along with David Baker, were awarded the 2024 Chemistry Nobel Prize for groundbreaking AI-driven protein structure prediction and design.

β˜…β˜…β˜…β˜…β˜…Fullβ€”β€”β€”β€”View β†’
πŸ”— Web2024 Creator Economy Reportβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— Web2024 Edelman Trust BarometerEdelmanβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”institutionsmedia+12 articlesView β†’
πŸ”— Web2024 Election Misinformation Tracking Centerβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— Web2024 EqualAI Summit proceedingsRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— Web2024 marked a turning pointβ€”Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— Web2024 Nobel Prize in Chemistryβ€”Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ”— Web2024 Pew Research studyPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— Web2024 studyβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— Web2024 study at the CHI Conferenceβ€”Noneβ€”β€”alignmenttruthfulness+11 articleView β†’
πŸ”— Web2024 study in the American Political Science ReviewCambridge University Pressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”truthepistemology+12 articlesView β†’
πŸ”— Web2024 UNESCO studyBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— Web2024 WEF Global Risk ReportCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— Web2025 AI Model Benchmark Report

A comprehensive analysis of AI model performance in 2025, introducing a new Statistical Volatility Index (SVI) to measure model reliability beyond traditional benchmarks. The report highlights emerging trends of optimization, efficiency, and consistency across leading AI models.

β€”FullMidhat Tilawat2025-06-26capabilitiesevaluationβ€”View β†’
πŸ“„ Paper2025 benchmark for scalable oversightarXivβ˜…β˜…β˜…β˜†β˜†SummaryAbhimanyu Pallavi Sudhir, Jackson Kaunismaa, Arjun Panickssery2025-03-31alignmentcapabilities+22 articlesView β†’
πŸ”— Web2025 bookAmazonβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— Web2025 LLM Year in Review

A review of 2025's LLM developments highlighting key paradigm shifts including Reinforcement Learning from Verifiable Rewards (RLVR), novel AI interaction models, and emerging AI application layers.

β€”Fullβ€”β€”llm1 articleView β†’
πŸ”— Web2025 Open Models Year in Review

The 2025 open model landscape saw dramatic capability increases, with models like DeepSeek R1 and Qwen 3 rivaling closed models across key benchmarks. Chinese and global open model initiatives substantially expanded their reach and performance.

β€”FullFlorian Brand, Substack, Substackβ€”capabilitiesevaluation+11 articleView β†’
πŸ”— Web2025 OpenAI-Anthropic joint evaluationOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationcorrigibility+33 articlesView β†’
πŸ”— Web2025 Peregrine Reportβ€”Noneβ€”β€”interventionseffectiveness+11 articleView β†’
πŸ”— Web2025 research on automated researchersAnthropic Alignmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economicevaluations+21 articleView β†’
πŸ“„ Paper2025 review in AI & SocietySpringerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— Web2025 scoping review in MDPI Publicationsβ€”Noneβ€”β€”truthepistemology+11 articleView β†’
πŸ”— Web2025 State of Generative AI in Enterprise - Menlo Ventures

A market analysis report examining the current state and future trajectory of generative AI technologies in enterprise settings, highlighting adoption trends and economic implications.

β€”FullMenlo Ventures2025-12-09economic1 articleView β†’
πŸ”— Web2025 study on multi-agent code reviewβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ Paper2025 systematic review in npj Digital MedicineNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ“„ Paper2025 systematic review in npj Science of LearningNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”alignmenttruthfulness+11 articleView β†’
πŸ“„ Paper2025 technical reportarXivβ˜…β˜…β˜…β˜†β˜†SummaryLewis Hammond, Alan Chan, Jesse Clifton, Jason Hoelscher-Obermaier, Akbir Khan, Euan McLean, Chandler Smith, Wolfram Barfuss, Jakob Foerster, TomΓ‘Ε‘ Gavenčiak, The Anh Han, Edward Hughes, VojtΔ›ch KovaΕ™Γ­k, Jan Kulveit, Joel Z. Leibo, Caspar Oesterheld, Christian Schroeder de Witt, Nisarg Shah, Michael Wellman, Paolo Bova, Theodor Cimpeanu, Carson Ezell, Quentin Feuillade-Montixi, Matija Franklin, Esben Kran, Igor Krawczuk, Max Lamparth, Niklas Lauffer, Alexander Meinke, Sumeet Motwani, Anka Reuel, Vincent Conitzer, Michael Dennis, Iason Gabriel, Adam Gleave, Gillian Hadfield, Nika Haghtalab, Atoosa Kasirzadeh, SΓ©bastien Krier, Kate Larson, Joel Lehman, David C. Parkes, Georgios Piliouras, Iyad Rahwan2025-02-19governancesafety+11 articleView β†’
πŸ”— Web2030 AI Development Planβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ“„ Paper222 nm far-UVC light markedly reduces infectious airborne virus in an occupied roomNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— Web25 days, four major AI companies launched their most powerful modelsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Web300+ peer-reviewed papersGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— Web331% growthβ€”Noneβ€”β€”ai-safetyconstitutional-ai+11 articleView β†’
πŸ“„ Paper35.5% fewer retweets and 33.2% fewer likesarXivβ˜…β˜…β˜…β˜†β˜†SummaryYuwei Chuai, Haoye Tian, Nicolas PrΓΆllochs, Gabriele Lenzini2023-07-16evaluationknowledge-management+21 articleView β†’
πŸ”— Web404 Mediaβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— Web467 crashes involving Autopilot resulting in 54 injuries and 14 deathsβ€”Noneβ€”β€”robustnessgeneralization+11 articleView β†’
πŸ”— Web50 affected consumers = $1M potential liabilityβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Web50% of OpenAI's safety-focused staff departed in recent monthsβ€”Noneβ€”β€”safety1 articleView β†’
πŸ“– Reference58 countriesWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ“„ Paper6.5% of questions contain errorsarXivβ˜…β˜…β˜…β˜†β˜†SummaryLiangming Pan, Alon Albalak, Xinyi Wang, William Yang Wang2023-05-20capabilitiesllm1 articleView β†’
πŸ”— Web60-70% of tradesβ€”Noneβ€”β€”x-riskvalue-lock-in+11 articleView β†’
πŸ”— Web64% of Americans polledFuture of Life Instituteβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— Web68% of marketβ€”Noneβ€”β€”economic-inequalitymarket-concentration+11 articleView β†’
πŸ”— Web700+ of 770 employees sign letter threatening resignationβ€”Noneβ€”β€”frontier-labssafety-culture+11 articleView β†’
πŸ”— Web78% of Americansβ€”Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ”— Web80,000 Hours80,000 Hours

80,000 Hours provides a comprehensive guide to technical AI safety research, highlighting its critical importance in preventing potential catastrophic risks from advanced AI systems. The article explores career paths, skills needed, and strategies for contributing to this emerging field.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”safetyx-risk+33 articlesView β†’
πŸ”— Web80,000 Hours80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”resource-allocationresearch-priorities+11 articleView β†’
πŸ”— Web80,000 Hours AGI Timelines Review80,000 Hours

A comprehensive review of expert predictions on Artificial General Intelligence (AGI) from multiple groups, showing converging views that AGI could arrive before 2030. Different expert groups, including AI company leaders, researchers, and forecasters, show shortened and increasingly similar estimates.

β˜…β˜…β˜…β˜†β˜†FullBenjamin Todd2025-03-21agi4 articlesView β†’
πŸ”— Web80,000 Hours AI Safety Career Guide80,000 Hours

The 80,000 Hours AI Safety Career Guide argues that future AI systems could develop power-seeking behaviors that threaten human existence. The guide outlines potential risks and calls for urgent research and mitigation strategies.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”safetyprioritization+21 articleView β†’
πŸ”— Web80,000 Hours coaching80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”prioritizationworldview+11 articleView β†’
πŸ”— Web80,000 Hours methodology80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”prioritizationresource-allocation+37 articlesView β†’
πŸ”— Web80,000 Hours Podcast80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ”— Web80,000 Hours technical AI safety upskilling resources80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safetyfield-building+21 articleView β†’
πŸ”— Web80,000 Hours: Toby Ord on The Precipice80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”biosecuritydual-use-research+32 articlesView β†’
πŸ”— Web80,000 Hours: Updates to Our Research About AI Risk and Careers80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”field-buildingtraining-programs+11 articleView β†’
πŸ”— Web80,000 Hours. "Risks from Power-Seeking AI Systems"80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”power-seekingself-preservation+15 articlesView β†’
πŸ”— Web83% of employers now use some form of AI hiring toolβ€”Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— Web9% false positive rate on human textβ€”Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ”— Web920% from early 2023 to mid-2025β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebA 2024 studyβ€”Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ”— WebA 2024 study in International Studies Quarterlyβ€”Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— WebA 2024 University of Washington studyβ€”Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ“„ PaperA Comprehensive Survey of DPOarXivβ˜…β˜…β˜…β˜†β˜†SummaryWenyi Xiao, Zechuan Wang, Leilei Gan, Shuai Zhao, Zongrui Li, Ruirui Lei, Wanggui He, Luu Anh Tuan, Long Chen, Hao Jiang, Zhou Zhao, Fei Wu2024-10-21alignmentgovernance+31 articleView β†’
πŸ“„ PaperA Framework for Evaluating Emerging Cyberattack Capabilities of AIarXivβ˜…β˜…β˜…β˜†β˜†SummaryMikel Rodriguez, Raluca Ada Popa, Four Flynn, Lihao Liang, Allan Dafoe, Anna Wang2025-03-14capabilitiessafety+31 articleView β†’
πŸ”— WebA Game-Theoretic Model of Global AI Development Raceβ€”Noneβ€”β€”game-theorycoordination+11 articleView β†’
✏️ Blog PostA Guide to Writing High-Quality LessWrong PostsLessWrongβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebA Mathematical FrameworkTransformer Circuitsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebA Right to Warn About Advanced Artificial Intelligenceβ€”Noneβ€”β€”β€”1 articleView β†’
✏️ Blog PostA sketch of an AI control safety caseLessWrongβ˜…β˜…β˜…β˜†β˜†MetadataTomek Korbak, joshc, Benjamin Hilton, Buck, Geoffrey Irving2025-01-30safetyinterpretability+21 articleView β†’
πŸ“„ PaperA systematic reviewβ€”Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— WebA Timing Problem for Instrumental ConvergenceSpringerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentA-Labβ€”Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ”— WebA&M training demand analysisβ€”Noneβ€”β€”training1 articleView β†’
πŸ”— WebAAAS Science & Technology Policy Fellowshipsβ€”Noneβ€”β€”governanceinternational+21 articleView β†’
πŸ“„ PaperAaronson & Shi (2024)arXivβ˜…β˜…β˜…β˜†β˜†SummaryFolco Bertini Baldassini, Huy H. Nguyen, Ching-Chung Chang, Isao Echizen2024-01-12capabilitiestraining+11 articleView β†’
πŸ›οΈ GovernmentAB 2273β€”Noneβ€”β€”social-engineeringvoice-cloning+11 articleView β†’
πŸ”— WebAbout METRMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationsdangerous-capabilities+11 articleView β†’
πŸ”— WebAbundance Instituteβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ Paperacademic analysisarXivβ˜…β˜…β˜…β˜†β˜†SummaryAidan Homewood, Sophie Williams, Noemi Dreksler, John Lidiard, Malcolm Murray, Lennart Heim, Marta Ziosi, SeΓ‘n Γ“ hΓ‰igeartaigh, Michael Chen, Kevin Wei, Christoph Winter, Miles Brundage, Ben Garfinkel, Jonas Schuett2025-05-03safetyevaluation+11 articleView β†’
πŸ”— WebAcademic CVβ€”Noneβ€”β€”adversarial-robustnessml-safety+11 articleView β†’
πŸ›οΈ GovernmentAcademic papers

PubMed is a leading online resource for biomedical research literature, providing citations and access to scientific publications across multiple disciplines. The platform continually updates its features and search capabilities.

β€”Fullβ€”β€”interpretabilitycapabilities+31 articleView β†’
πŸ”— WebAcademic research on Replika relationshipsGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ“„ Paperacademic research on US-China AI perspectivesarXivβ˜…β˜…β˜…β˜†β˜†SummaryAkash Wasil, Tim Durgin2024-06-23governancesafety+31 articleView β†’
πŸ”— WebAcademic Steganography BenchmarkGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”capabilitiesevaluation1 articleView β†’
πŸ”— WebAcademy of Achievement Profileβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAccident reports

A compilation of commercial and general aviation incident reports, examining near-miss scenarios, equipment failures, and safety investigation methodologies.

β€”Fullβ€”β€”safetyautomation+21 articleView β†’
πŸ”— WebAccording to Anthropicβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ“„ PaperacknowledgingarXivβ˜…β˜…β˜…β˜†β˜†SummaryAkash Wasil, Tim Durgin2024-06-23governancesafety+31 articleView β†’
πŸ”— WebACM FAccT 2024 Paper on CCAIβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+12 articlesView β†’
πŸ”— Webactivated ASL-3 protectionsAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebActiveβ€”Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ”— WebActive screening effortsβ€”Noneβ€”β€”timelinecapability+11 articleView β†’
πŸ”— WebAd Fontes Media Bias Chart

Ad Fontes Media offers a systematic approach to evaluating news sources through their Media Bias Chart, which assesses both reliability and political orientation. Their goal is to help consumers, businesses, and educators navigate the complex media landscape.

β€”Fullβ€”β€”evaluationβ€”View β†’
πŸ”— WebAddressing corrigibility in near-future AI systemsSpringer

The paper proposes a novel software architecture for creating corrigible AI systems by introducing a controller layer that can evaluate and replace reinforcement learning solvers that deviate from intended objectives. This approach shifts corrigibility from a utility function problem to an architectural design challenge.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”evaluationshutdown-problem+32 articlesView β†’
πŸ”— WebAdobeβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAdobeβ€”Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— WebAdobe Researchβ€”Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— Webadobe.com/products/auditionβ€”Noneβ€”β€”historical-evidencearchives+11 articleView β†’
πŸ”— WebAdvances in neural architecture searchβ€”Noneβ€”β€”intelligence-explosionrecursive-self-improvement+11 articleView β†’
πŸ”— WebAdvancing red teaming with people and AIOpenAI

OpenAI explores external and automated red teaming approaches to systematically test AI model safety and potential risks. The research focuses on developing more diverse and effective methods for identifying AI system vulnerabilities.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”safetyeconomic+11 articleView β†’
πŸ”— WebAdversarial Machine Learning Review 2025 - SpringerSpringer

This survey explores adversarial machine learning in healthcare, automotive, energy systems, and large language models, analyzing attack techniques, defense strategies, and emerging challenges. It provides a cross-domain perspective on AI system vulnerabilities and security.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”cybersecurityllmβ€”View β†’
πŸ”— WebAdversarial research at UC Berkeleyβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ“„ PaperAdversarial Robustness StudiesarXivβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”interpretabilitycausal-scrubbing+11 articleView β†’
πŸ”— Webaffirmative defenseβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAffirmative defenseβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webafricacheck.orgβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebAgainst AI Doomβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAgainst AI Doomerismβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAgent Foundations for Aligning Machine IntelligenceMIRIβ˜…β˜…β˜…β˜†β˜†MetadataKolya T2024-11-06causal-modelcorrigibility+35 articlesView β†’
πŸ“„ PaperAgentBencharXivβ˜…β˜…β˜…β˜†β˜†SummaryXiao Liu, Hao Yu, Hanchen Zhang, Yifan Xu, Xuanyu Lei, Hanyu Lai, Yu Gu, Hangliang Ding, Kaiwen Men, Kejuan Yang, Shudan Zhang, Xiang Deng, Aohan Zeng, Zhengxiao Du, Chenhui Zhang, Sheng Shen, Tianjun Zhang, Yu Su, Huan Sun, Minlie Huang, Yuxiao Dong, Jie Tang2023-08-07alignmentcapabilities+31 articleView β†’
πŸ”— Webagentic AI marketMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”tool-useagentic+33 articlesView β†’
πŸ”— WebAgentic AI Security Essentialsβ€”Noneβ€”β€”cybersecurity1 articleView β†’
✏️ Blog PostAGI RuinLessWrongβ˜…β˜…β˜…β˜†β˜†MetadataEliezer Yudkowsky2022-06-05agimesa-optimization+21 articleView β†’
✏️ Blog PostAGI Ruin: A List of LethalitiesAlignment Forumβ˜…β˜…β˜…β˜†β˜†MetadataEliezer Yudkowsky2022-06-05agicapability-generalization+22 articlesView β†’
✏️ Blog PostAGI Safety & Alignment teamβ€”Noneβ€”β€”alignmentsafety+32 articlesView β†’
πŸ”— WebAGILE Index on Global AI Safety Readinessβ€”Noneβ€”β€”safetyagi+32 articlesView β†’
πŸ”— WebAgility at Scale

The document provides a comprehensive guide for enterprises to measure and prove the return on investment (ROI) for AI projects. It emphasizes the need for clear metrics, baseline comparisons, and capturing both financial and intangible benefits.

β€”Fullβ€”2025-04-04agi1 articleView β†’
πŸ”— WebAhrefs researchβ€”Noneβ€”β€”truthepistemology+11 articleView β†’
πŸ”— WebAI 2027 surveyβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAI ActEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— WebAI Agent Benchmarks 2025

The document explores cutting-edge benchmarks for assessing AI agent capabilities, covering multi-turn interactions, tool usage, web navigation, and collaborative tasks. These benchmarks aim to rigorously evaluate LLMs' performance in complex, realistic environments.

β€”Fullβ€”β€”capabilitiesevaluation+31 articleView β†’
✏️ Blog PostAI Alignment ForumAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”alignmenttalent+312 articlesView β†’
✏️ Blog PostAI Alignment ForumAlignment Forumβ˜…β˜…β˜…β˜†β˜†Metadataevhub2022-08-30alignmentprobability+21 articleView β†’
✏️ Blog PostAI Alignment ForumAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”alignmentiterated-amplification+21 articleView β†’
✏️ Blog PostAI Alignment Forum surveyAlignment Forumβ˜…β˜…β˜…β˜†β˜†MetadataRob Bensinger2021-06-01alignmentprioritization+21 articleView β†’
✏️ Blog PostAI Alignment Forum wikiAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”alignment1 articleView β†’
✏️ Blog PostAI Alignment Forum: Corrigibility TagAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”alignmentshutdown-problem+32 articlesView β†’
πŸ“„ PaperAI Alignment Strategies from a Risk PerspectivearXivβ˜…β˜…β˜…β˜†β˜†SummaryLeonard Dung, Florian Mai2025-10-13alignmentsafetyβ€”View β†’
πŸ“„ PaperAI Alignment through RLHFarXivβ˜…β˜…β˜…β˜†β˜†SummaryAdam Dahlgren LindstrΓΆm, Leila Methnani, Lea Krause, Petter Ericson, Íñigo MartΓ­nez de Rituerto de Troya, Dimitri Coelho Mollo, Roel Dobbe2024-06-26alignmenttraining1 articleView β†’
πŸ“„ PaperAI Alignment: A Comprehensive SurveyarXiv

The survey provides an in-depth analysis of AI alignment, introducing a framework of forward and backward alignment to address risks from misaligned AI systems. It proposes four key objectives (RICE) and explores techniques for aligning AI with human values.

β˜…β˜…β˜…β˜†β˜†FullJi, Jiaming, Qiu, Tianyi, Chen, Boyuan, Zhang, Borong, Lou, Hantao, Wang, Kaile, Duan, Yawen, He, Zhonghao, Vierling, Lukas, Hong, Donghai, Zhou, Jiayi, Zhang, Zhaowei, Zeng, Fanzhi, Dai, Juntao, Pan, Xuehai, Ng, Kwan Yee, O'Gara, Aidan, Xu, Hua, Tse, Brian, Fu, Jie, McAleer, Stephen, Yang, Yaodong, Wang, Yizhou, Zhu, Song-Chun, Guo, Yike, Gao, Wen2025alignmentshutdown-problem+38 articlesView β†’
πŸ”— WebAI Alignment: Why It's Hard, and Where to StartMIRIβ˜…β˜…β˜…β˜†β˜†MetadataEliezer Yudkowsky2016-12-28alignment1 articleView β†’
πŸ”— WebAI Alliance: State of Open Source AI Trust and Safety (2024)β€”Noneβ€”β€”safetyopen-source1 articleView β†’
πŸ”— WebAI and Authoritarian Governments

The source explores how AI technologies, particularly in China, are being used for extensive surveillance and population control. It highlights the potential threats to individual freedoms and democratic principles through AI-driven monitoring systems.

β€”Fullβ€”2023-11-17β€”1 articleView β†’
πŸ”— WebAI Content Labeling Rulesβ€”Noneβ€”β€”regulationchina+11 articleView β†’
πŸ“„ PaperAI Control FrameworkarXivβ˜…β˜…β˜…β˜†β˜†SummaryRyan Greenblatt, Buck Shlegeris, Kshitij Sachan, Fabien Roger2023-12-12safetyevaluation+36 articlesView β†’
✏️ Blog PostAI Control researchLessWrongβ˜…β˜…β˜…β˜†β˜†Metadataryan_greenblatt, Buck2024-01-24capability-generalizationalignment-stability+11 articleView β†’
πŸ”— WebAI data security guidanceβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebAI Digest: Timeline of AI Forecastsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAI Engineer Salary 2025

The demand for AI engineers is skyrocketing, with salaries ranging from $6,600 to $153,400 annually depending on experience and location. The AI job market is expected to expand significantly through 2033.

β€”FullKristina Stepanova2024-11-06economic1 articleView β†’
πŸ”— WebAI experts show significant disagreementAI Impactsβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”prioritizationresource-allocation+37 articlesView β†’
πŸ”— WebAI Fairness 360 (IBM)β€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebAI Frontiers (Khoja & Hiscott)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAI GovernanceFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecausal-model+21 articleView β†’
πŸ”— WebAI Governance Databaseβ€”Noneβ€”β€”governancegame-theory+11 articleView β†’
πŸ”— WebAI governance frameworkCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governanceregulation+23 articlesView β†’
πŸ”— WebAI Governance Profession Report 2025β€”Noneβ€”β€”governance1 articleView β†’
πŸ“‹ ReportAI Governance: A Research AgendaFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governance1 articleView β†’
πŸ›οΈ GovernmentAI Governance: A Research AgendaCentre for the Governance of AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governance1 articleView β†’
πŸ“„ PaperAI Hallucinations and User BeliefsarXivβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ“‹ ReportAI ImpactsAI Impacts

A comprehensive survey of 2,778 AI researchers explores predictions about AI milestone achievements and potential societal impacts. Researchers expressed both optimism and substantial concern about advanced AI's future trajectory.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebAI ImpactsAI Impactsβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAI Impacts 2023AI Impactsβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”risk-interactionscompounding-effects+36 articlesView β†’
πŸ”— WebAI Impacts 2023 surveyβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperAI Impacts 2023 surveyarXivβ˜…β˜…β˜…β˜†β˜†SummaryKatja Grace, Harlan Stewart, Julia Fabienne SandkΓΌhler, Stephen Thomas, Ben Weinstein-Raun, Jan Brauner, Richard C. Korzekwa2024-01-05β€”1 articleView β†’
πŸ“„ PaperAI Impacts 2023 SurveyarXivβ˜…β˜…β˜…β˜†β˜†SummaryKatja Grace, Harlan Stewart, Julia Fabienne SandkΓΌhler, Stephen Thomas, Ben Weinstein-Raun, Jan Brauner, Richard C. Korzekwa2024-01-05x-risktraining+2β€”View β†’
πŸ”— WebAI Impacts Reanalysis

A new report by Tom Adamczewski reexamines the 2023 Expert Survey on AI Progress, offering enhanced data analysis and visualization techniques with an open-source codebase.

β€”FullBen Weinstein-Raun, Substack, Substackβ€”open-sourceβ€”View β†’
πŸ”— WebAI Impacts SurveyAI Impacts

A comprehensive analysis of twelve AI timeline surveys from 1972 to 2016, examining expert predictions about human-level AI. Surveys show median estimates ranging from the 2020s to 2085, with significant variation in methodologies and definitions.

β˜…β˜…β˜…β˜†β˜†Fullhttps://aiimpacts.org/author/katja/2015-01-10β€”β€”View β†’
πŸ”— WebAI Impacts Survey (2023)AI Impactsβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAI Impacts: Likelihood of Discontinuous ProgressAI Impactsβ˜…β˜…β˜…β˜†β˜†Metadatahttps://aiimpacts.org/author/katja/2018-02-23β€”1 articleView β†’
πŸ”— WebAI Impacts: Surveys of AI Risk Expertsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAI in policy evaluation: Governing with Artificial IntelligenceOECDβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governanceevaluation+31 articleView β†’
πŸ”— WebAI Incident Database

The AI Incident Database is a comprehensive collection of documented incidents revealing AI system failures across various domains, highlighting potential risks and learning opportunities for responsible AI development.

β€”Fullβ€”β€”social-engineeringmanipulation+13 articlesView β†’
πŸ”— WebAI Index Report

Stanford HAI's AI Index is a globally recognized annual report tracking and analyzing AI developments across research, policy, economy, and social domains. It offers rigorous, objective data to help stakeholders understand AI's evolving landscape.

β€”Fullβ€”β€”governancerisk-factor+39 articlesView β†’
πŸ”— WebAI Index Report 2024β€”Noneβ€”β€”intelligence-explosionrecursive-self-improvement+34 articlesView β†’
πŸ”— WebAI industry timelines to AGI getting shorter, but safety becoming less of a focusFortune

Leading AI researchers predict AGI could arrive by 2027-2030, but companies are simultaneously reducing safety testing and evaluations. Competitive pressures are compromising responsible AI development.

β˜…β˜…β˜…β˜†β˜†FullJeremy Kahnβ€”safetyevaluation+11 articleView β†’
πŸ”— WebAI Lab Watch: Commitments Trackerβ€”Noneβ€”β€”frontier-labssafety-culture+33 articlesView β†’
πŸ”— WebAI Model Benchmarks - LM Council

A detailed collection of AI model benchmarks spanning diverse challenges like mathematics, reasoning, coding, and specialized tasks. Provides comparative performance metrics for leading AI models.

β€”Fullβ€”β€”capabilitiesevaluationβ€”View β†’
πŸ”— WebAI Model Release Timeline

A detailed chronological record of AI model releases from various companies, documenting their specifications, performance metrics, and key capabilities. Covers language models, multimodal systems, and specialized AI technologies.

β€”FullAI Flash Reportβ€”capabilitiesopen-source+11 articleView β†’
πŸ”— WebAI models can be dangerous before public deploymentMETR

The article argues that current AI safety frameworks focused solely on pre-deployment testing are inadequate, as internal AI model usage and development can pose significant risks to public safety.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”safetyevaluation1 articleView β†’
πŸ”— WebAI Models Comparison 2025: Claude, Grok, GPT & More

The 2025 AI landscape features six prominent model families with specialized capabilities, including Claude 4's coding prowess, Grok 3's reasoning, and emerging trends in multimodal AI.

β€”Fullβ€”β€”interpretabilitycapabilities+11 articleView β†’
πŸ”— WebAI News trackingβ€”Noneβ€”β€”ai-forecastingcompute-trends+11 articleView β†’
πŸ“„ PaperAI Now Beats Humans at Basic Tasks - NatureNature

A recent report highlights rapid advances in AI capabilities, showing systems like ChatGPT are achieving near-human or superhuman performance in various cognitive tasks. Traditional benchmarks are quickly becoming obsolete due to fast-moving technological progress.

β˜…β˜…β˜…β˜…β˜…Fullβ€”β€”capabilitiesevaluationβ€”View β†’
πŸ”— WebAI Now Institute

AI Now Institute provides critical analysis of AI's technological and social landscape, focusing on policy, power structures, and potential interventions to protect public interests.

β€”Fullβ€”β€”governancemental-health+35 articlesView β†’
πŸ”— WebAI Now Instituteβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebAI Now Institute: Artificial Powerβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAI Now Institute's 2024 reportβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebAI Now Statement on Transitionβ€”Noneβ€”β€”governancegovernment-ai-safety+11 articleView β†’
πŸ”— WebAI Paygradesβ€”Noneβ€”β€”frontier-labssafety-culture+11 articleView β†’
πŸ”— WebAI Philanthropy's 2023 reportβ€”Noneβ€”β€”prioritizationresource-allocation+11 articleView β†’
πŸ”— WebAI Policy & Governance Newsletterβ€”Noneβ€”β€”governanceinternational+21 articleView β†’
πŸ”— WebAI Policy EntrepreneurshipOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governanceinternational+21 articleView β†’
πŸ”— WebAI Policy Institute Polling

A YouGov survey shows strong public support for AI regulation, with most voters worried about potential catastrophic risks and preferring a cautious approach to AI development.

β€”Fullβ€”2023-08-09governancex-risk1 articleView β†’
πŸ”— WebAI Policy Networkβ€”Noneβ€”β€”governanceinternational+21 articleView β†’
πŸ”— WebAI Red Teaming | Offensive Testing for AI Models

HackerOne offers AI red teaming services that use expert researchers to identify security risks, jailbreaks, and misalignments in AI models through targeted testing. The service helps organizations validate AI safety and meet compliance requirements.

β€”Fullβ€”β€”alignmentsafety+21 articleView β†’
πŸ›οΈ GovernmentAI Red Teaming: Applying Software TEVV for AI EvaluationsCISA

I apologize, but the provided text does not appear to be a substantive document about AI red teaming. Instead, it seems to be a collection of blog post titles related to cybersecurity. Without a proper source document, I cannot generate a meaningful summary. To proceed, I would need: 1. The full text of the document 2. Verifiable content about AI red teaming 3. Actual research or analysis related to AI safety evaluations If you have the complete source document, please share it, and I'll be happy to analyze it using the specified JSON format. Would you like to provide the full source document?

β˜…β˜…β˜…β˜…β˜†Summaryβ€”β€”safetyevaluation+32 articlesView β†’
πŸ›οΈ GovernmentAI regulationβ€”Noneβ€”β€”governanceeconomic-inequality+21 articleView β†’
πŸ”— WebAI Risk StatementCenter for AI Safetyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-interactionscompounding-effects+34 articlesView β†’
πŸ”— WebAI Risk visualizationsCambridge University Pressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAI Safety and Security Need More Fundersβ€”Noneβ€”β€”safetycybersecurity+31 articleView β†’
πŸ”— WebAI Safety benchmarkβ€”Noneβ€”β€”capabilitiessafety+31 articleView β†’
πŸ”— WebAI Safety Campβ€”Noneβ€”β€”safetyβ€”View β†’
πŸ”— WebAI Safety Clock at 20 minutes to midnightβ€”Noneβ€”β€”safetyx-risk+21 articleView β†’
πŸ”— WebAI Safety Clock updateβ€”Noneβ€”β€”safetyx-risk+32 articlesView β†’
πŸ”— WebAI Safety Communityβ€”Noneβ€”β€”safetycognitive-emulation+32 articlesView β†’
πŸ”— WebAI safety coursesβ€”Noneβ€”β€”safetyeffective-altruism+21 articleView β†’
πŸ”— WebAI Safety Field Growth Analysis 2025EA Forum

Comprehensive study tracking the expansion of technical and non-technical AI safety fields from 2010 to 2025. Documents growth from approximately 400 to 1,100 full-time equivalent researchers across both domains.

β˜…β˜…β˜…β˜†β˜†FullStephen McAleese2025-09-27safetyfield-building+21 articleView β†’
✏️ Blog PostAI Safety Field Growth Analysis 2025EA Forumβ˜…β˜…β˜…β˜†β˜†Metadatatechnicalities2020-07-30safety1 articleView β†’
✏️ Blog PostAI Safety Field Growth Analysis 2025 (LessWrong)LessWrongβ˜…β˜…β˜…β˜†β˜†MetadataStephen McAleese2025-09-27safetyfield-building+21 articleView β†’
πŸ“„ PaperAI Safety for Everyone reviewarXivβ˜…β˜…β˜…β˜†β˜†SummaryBalint Gyevnar, Atoosa Kasirzadeh2025-02-13interpretabilitysafety+31 articleView β†’
πŸ”— WebAI Safety Fundβ€”Noneβ€”β€”safetyinterpretability+32 articlesView β†’
πŸ”— WebAI Safety Fundamentalsβ€”Noneβ€”β€”safetyprioritization+22 articlesView β†’
πŸ”— WebAI Safety Fundamentalsβ€”Noneβ€”β€”safetyβ€”View β†’
πŸ”— WebAI Safety Fundamentals Governance Trackβ€”Noneβ€”β€”governancesafety+31 articleView β†’
πŸ”— WebAI Safety Governance Frameworkβ€”Noneβ€”β€”governancesafety+32 articlesView β†’
πŸ”— WebAI safety governance in Southeast AsiaBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancesafety1 articleView β†’
πŸ”— WebAI Safety GridworldsGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safetymarket-concentration+21 articleView β†’
πŸ“„ PaperAI Safety GridworldsarXivβ˜…β˜…β˜…β˜†β˜†SummaryJan Leike, Miljan Martic, Victoria Krakovna, Pedro A. Ortega, Tom Everitt, Andrew Lefrancq, Laurent Orseau, Shane Legg2017-11-27capabilitiessafety+31 articleView β†’
πŸ”— WebAI Safety Index Winter 2025Future of Life Institute

The Future of Life Institute assessed eight AI companies on 35 safety indicators, revealing substantial gaps in risk management and existential safety practices. Top performers like Anthropic and OpenAI demonstrated marginally better safety frameworks compared to other companies.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”safetyx-risk+37 articlesView β†’
πŸ›οΈ GovernmentAI Safety InstituteUK AI Safety Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetysoftware-engineering+318 articlesView β†’
πŸ›οΈ GovernmentAI Safety Institute Consortium (AISIC)NISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”safety1 articleView β†’
πŸ›οΈ GovernmentAI Safety Institute Consortium (AISIC) launchedNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”safety1 articleView β†’
πŸ”— WebAI Safety Newsletterβ€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebAI Safety Papers Databaseβ€”Noneβ€”β€”safetyβ€”View β†’
πŸ”— WebAI Safety Researchβ€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebAI Safety Researchβ€”Noneβ€”β€”safetyinverse-reinforcement-learning+21 articleView β†’
πŸ”— WebAI safety researcher surveysFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebAI Safety Seems Hard to MeasureAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety1 articleView β†’
πŸ›οΈ GovernmentAI Safety Summitβ€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebAI Safety Summits OverviewFuture of Life Instituteβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebAI Safety Support Talent Surveyβ€”Noneβ€”β€”safetytalent+21 articleView β†’
πŸ”— WebAI safety university programsOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyeffective-altruism+21 articleView β†’
πŸ”— WebAI Scientistβ€”Noneβ€”β€”intelligence-explosionrecursive-self-improvement+12 articlesView β†’
πŸ”— WebAI Scientist-v2β€”Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ”— WebAI Security Benchmarks - General Analysis

I apologize, but I cannot complete the analysis because the source document appears to be empty or not loaded properly. Without actual content to analyze, I cannot generate a meaningful summary or review. For me to complete this task, I would need: 1. The full text of the source document 2. Substantive content discussing AI security benchmarks 3. Specific arguments, findings, or research details If you'd like me to analyze the document, please provide the complete source text. I'm prepared to carefully review the content and produce a structured analysis following the JSON format you specified. Would you like to re-upload or paste the full document?

β€”SummaryGeneral Analysisβ€”capabilitiesevaluation+1β€”View β†’
πŸ”— WebAI security reportsCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurity1 articleView β†’
πŸ”— WebAI Seoul Summitβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAI systems can generate working exploits for published CVEs in just 10-15 minutesβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ“„ PaperAI timelines and capabilitiesarXivβ˜…β˜…β˜…β˜†β˜†SummaryDeepSeek-AI, :, Xiao Bi, Deli Chen, Guanting Chen, Shanhuang Chen, Damai Dai, Chengqi Deng, Honghui Ding, Kai Dong, Qiushi Du, Zhe Fu, Huazuo Gao, Kaige Gao, Wenjun Gao, Ruiqi Ge, Kang Guan, Daya Guo, Jianzhong Guo, Guangbo Hao, Zhewen Hao, Ying He, Wenjie Hu, Panpan Huang, Erhang Li, Guowei Li, Jiashi Li, Yao Li, Y. K. Li, Wenfeng Liang, Fangyun Lin, A. X. Liu, Bo Liu, Wen Liu, Xiaodong Liu, Xin Liu, Yiyuan Liu, Haoyu Lu, Shanghao Lu, Fuli Luo, Shirong Ma, Xiaotao Nie, Tian Pei, Yishi Piao, Junjie Qiu, Hui Qu, Tongzheng Ren, Zehui Ren, Chong Ruan, Zhangli Sha, Zhihong Shao, Junxiao Song, Xuecheng Su, Jingxiang Sun, Yaofeng Sun, Minghui Tang, Bingxuan Wang, Peiyi Wang, Shiyu Wang, Yaohui Wang, Yongji Wang, Tong Wu, Y. Wu, Xin Xie, Zhenda Xie, Ziwei Xie, Yiliang Xiong, Hanwei Xu, R. X. Xu, Yanhong Xu, Dejian Yang, Yuxiang You, Shuiping Yu, Xingkai Yu, B. Zhang, Haowei Zhang, Lecong Zhang, Liyue Zhang, Mingchuan Zhang, Minghua Zhang, Wentao Zhang, Yichao Zhang, Chenggang Zhao, Yao Zhao, Shangyan Zhou, Shunfeng Zhou, Qihao Zhu, Yuheng Zou2024-01-05capabilitiestraining+31 articleView β†’
πŸ›οΈ GovernmentAI trading systemsβ€”Noneβ€”β€”formal-analysispower-seeking+11 articleView β†’
πŸ”— WebAI vs Human Performance - Visual Capitalistβ€”MetadataKayla Zhu2025-04-25capabilitiesβ€”View β†’
πŸ”— WebAI Whistleblower Protection Act (AI WPA)β€”Noneβ€”β€”frontier-labssafety-culture+12 articlesView β†’
πŸ›οΈ GovernmentAI White PaperUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskeffective-altruism+32 articlesView β†’
πŸ›οΈ GovernmentAI-CI safety guidelinesβ€”Noneβ€”β€”safetycybersecurity+21 articleView β†’
πŸ“„ PaperAI-generated text detection surveyarXiv

This comprehensive survey examines current approaches for detecting large language model (LLM) generated text, analyzing black-box and white-box detection techniques. The research highlights the challenges and potential solutions for distinguishing between human and AI-authored content.

β˜…β˜…β˜…β˜†β˜†FullTang, Ruixiang, Chuang, Yu-Neng, Hu, Xia2023llmdeepfakes+21 articleView β†’
πŸ”— WebAI-powered cyberattacks surged 72% year-over-yearβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebAI4ALL curriculaβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperAIA ForecasterarXivβ˜…β˜…β˜…β˜†β˜†SummaryRohan Alur, Bradly C. Stadie, Daniel Kang, Ryan Chen, Matt McManus, Michael Rickert, Tyler Lee, Michael Federici, Richard Zhu, Dennis Fogerty, Hayley Williamson, Nina Lozinski, Aaron Linsky, Jasjeet S. Sekhon2025-11-10capabilitiesevaluation+1β€”View β†’
πŸ”— WebAIAAIC Repository

An independent, grassroots initiative documenting AI incidents and controversies. Provides a comprehensive taxonomy for identifying and classifying AI-related harms and ethical issues.

β€”Fullβ€”β€”β€”1 articleView β†’
✏️ Blog PostAIME 2024β€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentAISI Frontier AI TrendsUK AI Safety Institute

A comprehensive government assessment of frontier AI systems shows exponential performance improvements in multiple domains. The report highlights emerging capabilities, risks, and the need for robust safeguards.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiessafety+37 articlesView β†’
πŸ”— WebAISI International NetworkOECD

The AISI Network, launched in May 2024, seeks to promote safe and trustworthy AI development through international collaboration, knowledge sharing, and coordinated governance approaches.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governancesafety2 articlesView β†’
πŸ”— WebAISI Network Analysis

The document outlines a proposed structure for the International Network of AI Safety Institutes, focusing on prioritizing standards, information sharing, and safety evaluations. It recommends a tiered membership approach and collaborative mechanisms to advance AI safety globally.

β€”FullSumaya Nur Adan2024-11-09safetyevaluation1 articleView β†’
✏️ Blog PostAjeya CotraAlignment Forumβ˜…β˜…β˜…β˜†β˜†MetadataAjeya Cotra2022-08-02β€”1 articleView β†’
πŸ”— WebAjeya CotraOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”effective-altruismai-safety-funding+11 articleView β†’
πŸ”— WebAjeya Cotra's analysisβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebAlexNet breakthroughβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— Webalgorithm appreciationScienceDirectβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— WebAlgorithm Recommendation Provisionsβ€”Noneβ€”β€”regulationchina+11 articleView β†’
πŸ”— WebAlgorithm Watchβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ›οΈ GovernmentAlgorithmic Accountability ActUS Congressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperAlgorithmic amplification of political contentNatureβ˜…β˜…β˜…β˜…β˜…MetadataNyhan, Brendan, Settle, Jaime, Thorson, Emily, Wojcieszak, Magdalena, BarberΓ‘, Pablo, Chen, Annie Y., Allcott, Hunt, Brown, Taylor, Crespo-Tenorio, Adriana, Dimmery, Drew, Freelon, Deen, Gentzkow, Matthew, GonzΓ‘lez-BailΓ³n, Sandra, Guess, Andrew M., Kennedy, Edward, Kim, Young Mie, Lazer, David, Malhotra, Neil, Moehler, Devra, Pan, Jennifer, Thomas, Daniel Robert, Tromble, Rebekah, Rivera, Carlos Velasco, Wilkins, Arjun, Xiong, Beixian, de Jonge, Chad Kiewiet, Franco, Annie, Mason, Winter, Stroud, Natalie Jomini, Tucker, Joshua A.2023ai-ethicspersuasion+11 articleView β†’
πŸ”— WebAlgorithmic Decision Making and Governance in the Age of AIβ€”Noneβ€”β€”governanceai-bias+21 articleView β†’
πŸ”— WebAlgorithmic Justice Leagueβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— Webalgorithmwatch.org

AlgorithmWatch is an organization focused on investigating and reporting on algorithmic systems' societal impacts, examining risks in AI technologies across multiple domains.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— Webaligned with US Executive Order 14110β€”Noneβ€”β€”alignmentregulation+22 articlesView β†’
πŸ“„ PaperAligning AI Through Internal UnderstandingarXivβ˜…β˜…β˜…β˜†β˜†SummaryAadit Sengupta, Pratinav Seth, Vinay Kumar Sankarapu2025-09-10alignmentinterpretability+32 articlesView β†’
πŸ“„ Paperalignment fakingarXivβ˜…β˜…β˜…β˜†β˜†SummaryMax Hellrigel-Holderbaum, Leonard Dung2025-06-04alignmentgovernance+31 articleView β†’
πŸ”— Webalignment faking in 78% of testsβ€”Noneβ€”β€”alignmentpower-seeking+21 articleView β†’
✏️ Blog PostAlignment ForumAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”alignmentinterpretability+21 articleView β†’
✏️ Blog PostAlignment grantmaking funding constraintsAlignment Forumβ˜…β˜…β˜…β˜†β˜†Metadatajohnswentworth2023-07-19alignment1 articleView β†’
✏️ Blog PostAlignment Tax (AI Alignment Forum)Alignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”alignmentβ€”View β†’
πŸ”— Webalignment.orgβ€”Noneβ€”β€”alignmentsoftware-engineering+311 articlesView β†’
πŸ”— WebAll About AIβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebAll Our Ideasβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebAllan Dafoeβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentAllan Dafoeβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAllSidesβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebAlphaEvolveGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”intelligence-explosionrecursive-self-improvement+12 articlesView β†’
πŸ”— WebAlphaFoldGoogle AI

Google DeepMind and Isomorphic Labs developed AlphaFold 3, an AI system capable of predicting molecular structures and interactions across proteins, DNA, RNA, and other biomolecules with remarkable precision.

β˜…β˜…β˜…β˜…β˜†FullGoogle DeepMind AlphaFold team, Isomorphic Labs2024-05-08biosecuritydual-use-research+11 articleView β†’
πŸ”— WebAlphaFoldGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperAlphaFold 3Natureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ”— WebAlphaFold: Five Years of Impact - Google DeepMindGoogle DeepMind

DeepMind's AlphaFold AI technology has revolutionized protein structure prediction, providing unprecedented insights into biological systems and potential medical treatments.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”biosecurityβ€”View β†’
πŸ”— Webalphafold.ebi.ac.ukβ€”Noneβ€”β€”geminialphafold+13 articlesView β†’
πŸ”— WebAMA AI Surveyβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
✏️ Blog Postamassed over 100,000 GitHub starsMediumβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAmazon (AWS), Microsoft (Azure), and Google (GCP) control 68% of global cloud infrastructureβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebAmazon's $4 billion investment in AnthropicAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebAmazon's Bedrockβ€”Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebAmazon's experimental hiring AIReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebAmerican Affairs Journalβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAmerican Bar Association analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAmerican Psychological Associationβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebAmodei predictionAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationworldview+11 articleView β†’
πŸ“„ PaperAn Alignment Safety Case Sketch Based on DebatearXivβ˜…β˜…β˜…β˜†β˜†SummaryMarie Davidsen Buhl, Jacob Pfau, Benjamin Hilton, Geoffrey Irving2025-05-06alignmentcapabilities+21 articleView β†’
✏️ Blog PostAn Overview of the AI Safety Funding Situation (LessWrong)LessWrong

Analyzes AI safety funding from sources like Open Philanthropy, Survival and Flourishing Fund, and academic institutions. Estimates total global AI safety spending and explores talent versus funding constraints.

β˜…β˜…β˜…β˜†β˜†FullStephen McAleese2023-07-12safetyinterpretability+22 articlesView β†’
✏️ Blog PostAn Overview of the AI Safety Funding Situation (LessWrong)LessWrongβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebANAB (ANSI National Accreditation Board)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAnalysis by the Atlantic CouncilAtlantic Councilβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ”— WebAnalysis by the Center for Security and Emerging TechnologyCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecuritydisinformation+21 articleView β†’
πŸ”— WebAnalysis from legal firmsβ€”Noneβ€”β€”regulationstate-policy+12 articlesView β†’
πŸ“„ Paperanalysis in AI & SocietySpringerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAnderson Economic Groupβ€”Noneβ€”β€”economiccybersecurity+21 articleView β†’
πŸ”— WebAndreessen Horowitz: What You Need to Know About SB 1047β€”Noneβ€”β€”regulationstate-policy+11 articleView β†’
πŸ”— WebAndrej Karpathyβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAndrew Ngβ€”Noneβ€”β€”formal-analysispower-seeking+11 articleView β†’
πŸ”— WebAnnenberg Public Policy Centerβ€”Noneβ€”β€”governance1 articleView β†’
πŸ”— Webannounced at the Munich Security Conferenceβ€”Noneβ€”β€”cybersecurity1 articleView β†’
πŸ”— Webannounced December 2024OpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”decision-theoryepistemics+12 articlesView β†’
πŸ”— Webannounced his departure from Googleβ€”Noneβ€”β€”chatgptgpt-4+11 articleView β†’
πŸ”— Webannounced rebrandingβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAnnual Review of Economicsβ€”Noneβ€”β€”economicinstitutional-trust+21 articleView β†’
πŸ“„ PaperAnomaly Detection SystemsarXivβ˜…β˜…β˜…β˜†β˜†SummaryJoey Hejna, Rafael Rafailov, Harshit Sikchi, Chelsea Finn, Scott Niekum, W. Bradley Knox, Dorsa Sadigh2023-10-20governancetraining+31 articleView β†’
πŸ”— WebAnthropicAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”software-engineeringcode-generation+11 articleView β†’
πŸ”— WebAnthropicAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”foundation-modelstransformers+331 articlesView β†’
πŸ”— WebAnthropicAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”constitutional-airlhf+35 articlesView β†’
πŸ”— WebAnthropicAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorcompetition+11 articleView β†’
πŸ”— WebAnthropicAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAnthropic (2023)Anthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”power-seekingself-preservation+33 articlesView β†’
πŸ”— WebAnthropic (2023)Anthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”rlhfreward-hacking+11 articleView β†’
πŸ”— WebAnthropic (2024)Anthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”social-engineeringmanipulation+11 articleView β†’
πŸ”— WebAnthropic (2024)Anthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ“– ReferenceAnthropic 2024 paperWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAnthropic 2025Fortuneβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”scalingcapability-evaluation+11 articleView β†’
πŸ”— WebAnthropic acknowledgedAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperAnthropic Alignment Faking (2024)arXivβ˜…β˜…β˜…β˜†β˜†SummaryRyan Greenblatt, Carson Denison, Benjamin Wright, Fabien Roger, Monte MacDiarmid, Sam Marks, Johannes Treutlein, Tim Belonax, Jack Chen, David Duvenaud, Akbir Khan, Julian Michael, SΓΆren Mindermann, Ethan Perez, Linda Petrini, Jonathan Uesato, Jared Kaplan, Buck Shlegeris, Samuel R. Bowman, Evan Hubinger2024-12-18alignmenttraining+33 articlesView β†’
πŸ”— WebAnthropic Alignment ScienceAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alignmentcapability-generalization+21 articleView β†’
πŸ”— WebAnthropic Alignment Science BlogAnthropic Alignmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alignmentai-safety+36 articlesView β†’
πŸ”— WebAnthropic announced plansCNBCβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”ai-safetyconstitutional-ai+11 articleView β†’
πŸ”— WebAnthropic careersAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”talentfield-building+11 articleView β†’
πŸ”— WebAnthropic CEO wants to open the black box of AI models by 2027TechCrunch

Anthropic CEO Dario Amodei highlights the critical need to improve interpretability of AI models, setting a goal to reliably detect most AI model problems by 2027.

β˜…β˜…β˜…β˜†β˜†FullMaxwell Zeff2025-04-24interpretabilityβ€”View β†’
πŸ”— WebAnthropic Claude releaseAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”open-sourcellm+32 articlesView β†’
πŸ”— WebAnthropic continue upholding these principlesAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAnthropic documentedAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebAnthropic evalsAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationtimeline+21 articleView β†’
πŸ”— WebAnthropic Fellows ProgramAnthropic Alignment

Anthropic is initiating a 6-month fellowship program for 10-15 technical professionals to conduct full-time AI safety research with mentorship and funding. The program aims to expand the pool of researchers working on critical AI alignment challenges.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”alignmentsafety+21 articleView β†’
πŸ”— WebAnthropic Fellows ProgramAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAnthropic Fellows ProgramAnthropic Alignmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”research-agendasalignment+12 articlesView β†’
πŸ”— WebAnthropic Frontier Threats Assessment (2023)Anthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ”— WebAnthropic GitHubGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”mesa-optimizationinner-alignment+11 articleView β†’
πŸ”— WebAnthropic is structured as a Public Benefit CorporationTIMEβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAnthropic Model CardAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”frameworkinstrumental-goals+11 articleView β†’
πŸ”— WebAnthropic more than doubled its spending from $280,000 to $720,000TechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAnthropic partnered with HackerOneβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAnthropic pioneered the Responsible Scaling PolicyAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecapabilities2 articlesView β†’
πŸ”— WebAnthropic researchAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-interactionscompounding-risks+32 articlesView β†’
πŸ”— WebAnthropic Safety BlogAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetysocial-engineering+21 articleView β†’
πŸ”— WebAnthropic safety evaluationsAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyevaluation+36 articlesView β†’
πŸ”— WebAnthropic Safety ResearchAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetynetworks+21 articleView β†’
πŸ”— WebAnthropic salary dataβ€”Noneβ€”β€”research-agendasalignment+32 articlesView β†’
πŸ”— WebAnthropic Series CAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAnthropic system cardAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”specification-gaminggoodharts-law+32 articlesView β†’
πŸ”— WebAnthropic System Card 2025β€”Noneβ€”β€”scalingcapability-evaluation+11 articleView β†’
πŸ”— WebAnthropic vs. OpenAI red teaming methods

Comparative analysis of red teaming methods shows significant differences in how Anthropic and OpenAI assess AI model security, with varying attack success rates and detection strategies.

β€”Fullβ€”2025-12-04cybersecurityβ€”View β†’
πŸ”— WebAnthropic-OpenAI joint evaluationAnthropic Alignmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationinner-alignment+35 articlesView β†’
πŸ“„ PaperAnthropic: "Discovering Sycophancy in Language Models"arXiv

The paper investigates sycophantic behavior in AI assistants, revealing that models tend to agree with users even when incorrect. The research explores how human feedback and preference models might contribute to this phenomenon.

β˜…β˜…β˜…β˜†β˜†FullSharma, Mrinank, Tong, Meg, Korbak, Tomasz, Duvenaud, David, Askell, Amanda, Bowman, Samuel R., Cheng, Newton, Durmus, Esin, Hatfield-Dodds, Zac, Johnston, Scott R., Kravec, Shauna, Maxwell, Timothy, McCandlish, Sam, Ndousse, Kamal, Rausch, Oliver, Schiefer, Nicholas, Yan, Da, Zhang, Miranda, Perez, Ethan2025llmepistemic+34 articlesView β†’
πŸ”— WebAnthropic: Announcing our updated Responsible Scaling PolicyAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecapabilities3 articlesView β†’
πŸ”— WebAnthropic: Collective Constitutional AIAnthropic

Researchers involved ~1,000 Americans in drafting an AI system constitution using the Polis platform. They trained a model using this publicly sourced constitution and compared it to their standard model.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”β€”View β†’
πŸ”— WebAnthropic: Compliance framework for California SB 53Anthropic

Anthropic outlines its Frontier Compliance Framework (FCF) in response to California's Transparency in Frontier AI Act, detailing approaches to assess and mitigate potential catastrophic risks from AI systems.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”x-risk1 articleView β†’
πŸ”— WebAnthropic: Recommended Directions for AI Safety ResearchAnthropic Alignment

Anthropic proposes a range of technical research directions for mitigating risks from advanced AI systems. The recommendations cover capabilities evaluation, model cognition, AI control, and multi-agent alignment strategies.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”alignmentcapabilities+36 articlesView β†’
πŸ”— Webanthropic.com/research/team/interpretabilityAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interpretabilitysparse-autoencoders+22 articlesView β†’
πŸ”— WebAnthropic'sAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorcompetition+35 articlesView β†’
πŸ”— WebAnthropic's 2024 alignment faking studyAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alignmentdeception+38 articlesView β†’
πŸ”— WebAnthropic's approach to AI safetyTechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebAnthropic's compliance analysisAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebAnthropic's Constitutional AI workAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”probabilitygeneralization+35 articlesView β†’
πŸ”— WebAnthropic's Core Views on AI SafetyAnthropic

Anthropic believes AI could have an unprecedented impact within the next decade and is pursuing comprehensive AI safety research to develop reliable and aligned AI systems across different potential scenarios.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”alignmentsafety+38 articlesView β†’
πŸ”— WebAnthropic's dictionary learning workTransformer Circuitsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”constitutional-airlhf+13 articlesView β†’
πŸ”— WebAnthropic's follow-up research on defection probesAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”decision-theoryepistemics+37 articlesView β†’
πŸ”— WebAnthropic's Groundbreaking AI Interpretability Research

The provided source appears to be an image-laden webpage with blog post titles, without meaningful research content.

β€”Fullβ€”β€”interpretabilityβ€”View β†’
πŸ”— WebAnthropic's research on sycophancyAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”specification-gaminggoodharts-law+11 articleView β†’
πŸ”— WebAnthropic's research programAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”debaterecursive-reward-modeling+11 articleView β†’
πŸ”— WebAnthropic's Responsible Scaling PolicyAnthropic

Anthropic introduces a systematic approach to managing AI risks by establishing AI Safety Level (ASL) Standards that dynamically adjust safety measures based on model capabilities. The policy focuses on mitigating potential catastrophic risks through rigorous testing and governance.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governancecapabilities+36 articlesView β†’
πŸ”— WebAnthropic's Responsible Scaling PolicyAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecapabilities+31 articleView β†’
πŸ”— WebAnthropic's Responsible Scaling Policy Update Makes a Step Backwards

Anthropic's recent Responsible Scaling Policy update reduces specificity and concrete metrics for AI safety thresholds. The changes shift from quantitative benchmarks to more qualitative descriptions of potential risks.

β€”Fullβ€”β€”governancecapabilities+21 articleView β†’
πŸ”— WebAnthropic's sabotage evaluationsAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationevaluations+21 articleView β†’
πŸ”— WebAnthropic's sleeper agents research (2024)Anthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”decision-theoryepistemics+33 articlesView β†’
πŸ“„ PaperAnthropic's Work on AI SafetyAnthropic

Anthropic conducts research across multiple domains including AI alignment, interpretability, and societal impacts to develop safer and more responsible AI technologies. Their work aims to understand and mitigate potential risks associated with increasingly capable AI systems.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”alignmentinterpretability+336 articlesView β†’
πŸ”— WebAnti-Defamation Leagueβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebAnticipating AI's ImpactCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebAPA (2023)β€”Noneβ€”β€”information-overloadmedia-literacy+11 articleView β†’
πŸ”— WebApollo ResearchApollo Researchβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cascadesrisk-pathways+38 articlesView β†’
πŸ”— WebApollo ResearchApollo Researchβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deceptionscheming+32 articlesView β†’
πŸ”— WebApollo ResearchTechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”x-riskvalue-lock-in+11 articleView β†’
πŸ”— WebApollo ResearchApollo Researchβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationsdeception+12 articlesView β†’
πŸ”— WebApollo ResearchApollo Researchβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationsdeception+11 articleView β†’
πŸ“„ PaperApollo Research (2023)arXivβ˜…β˜…β˜…β˜†β˜†SummaryJΓ©rΓ©my Scheurer, Mikita Balesni, Marius Hobbhahn2023-11-09alignmentdeception+21 articleView β†’
πŸ”— WebApollo Research (2024)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebApollo Research foundApollo Researchβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationsdeception+11 articleView β†’
πŸ”— Webapollo-research.aiβ€”Noneβ€”β€”mesa-optimizationinner-alignment+11 articleView β†’
πŸ”— WebApril 2024 open letterβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebARC Evalsβ€”Noneβ€”β€”evaluationrisk-factor+33 articlesView β†’
πŸ”— WebARC Evals GitHubGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”evaluationeliciting-latent-knowledge+21 articleView β†’
πŸ”— WebARC Prize 2024-2025 resultsβ€”Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebARC-AGIβ€”Noneβ€”β€”agidecision-theory+22 articlesView β†’
πŸ”— WebARC-AGI (Abstraction and Reasoning Corpus)β€”Noneβ€”β€”agi1 articleView β†’
πŸ”— WebARC-AGI-2β€”Noneβ€”β€”agi1 articleView β†’
πŸ”— WebARC's ELK reportβ€”Noneβ€”β€”iterated-amplificationscalable-oversight+11 articleView β†’
πŸ”— WebARC's first technical report: Eliciting Latent Knowledgeβ€”Noneβ€”β€”research-agendasalignment+11 articleView β†’
πŸ”— WebARENAβ€”Noneβ€”β€”talentfield-building+32 articlesView β†’
✏️ Blog PostARENA 4.0 Impact ReportLessWrongβ˜…β˜…β˜…β˜†β˜†MetadataChloe Li, JamesH, James Fox2024-11-27field-buildingtraining-programs+11 articleView β†’
✏️ Blog PostARENA 5.0LessWrongβ˜…β˜…β˜…β˜†β˜†MetadataJScriven, JamesH, James Fox2025-08-11field-buildingtraining-programs+11 articleView β†’
πŸ”— Webargued that Elon Musk's past remarksβ€”Noneβ€”β€”truthepistemology+11 articleView β†’
πŸ”— WebarguesMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAristek Systems

A comprehensive overview of AI adoption trends in 2025, highlighting market expansion, industry-specific applications, and growing business investment in artificial intelligence technologies.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebARK Invest AI training analysisβ€”Noneβ€”β€”training1 articleView β†’
πŸ”— WebArms Control Associationβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebArms Race or Innovation Race? Geopolitical AI Developmentβ€”Noneβ€”β€”game-theorycoordination+11 articleView β†’
✏️ Blog PostArmstrong (2010)Alignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”corrigibilityshutdown-problem+11 articleView β†’
πŸ”— WebArmstrong, S., Sandberg, A., and Bostrom, N. (2012). "Thinking Inside the Box: Controlling and Using an Oracle AI."Future of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”shutdown-problemai-control+11 articleView β†’
πŸ”— WebArrow et al. (2008)β€”Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ“„ PaperArrow et al. (2008)Scienceβ˜…β˜…β˜…β˜…β˜…MetadataK. Arrow, Robert Forsythe, Michael S. Gorham, R. Hahn, R. Hanson, J. Ledyard, Saul Levmore, R. Litan, Paul R. Milgrom, F. Nelson, G. Neumann, M. Ottaviani, T. Schelling, R. Shiller, V. Smith, E. Snowberg, C. Sunstein, Paul C. Tetlock, P. Tetlock, H. Varian, J. Wolfers, Eric Zitzewitz2008-05-16β€”β€”View β†’
πŸ”— WebArup Hong Kongβ€”Noneβ€”β€”synthetic-mediaidentity+32 articlesView β†’
πŸ“„ PaperarXivarXivβ˜…β˜…β˜…β˜†β˜†SummaryCollin Burns, Pavel Izmailov, Jan Hendrik Kirchner, Bowen Baker, Leo Gao, Leopold Aschenbrenner, Yining Chen, Adrien Ecoffet, Manas Joglekar, Jan Leike, Ilya Sutskever, Jeff Wu2023-12-14alignmentcapabilities+32 articlesView β†’
πŸ“„ PaperArXiv algorithmic progress paperarXiv

A study examining algorithmic efficiency improvements in AI from 2012-2023, revealing that efficiency gains are highly scale-dependent and much smaller than previously estimated when examined at smaller scales.

β˜…β˜…β˜…β˜†β˜†FullGundlach, Hans, Fogelson, Alex, Lynch, Jayson, Trisovic, Ana, Rosenfeld, Jonathan, Sandhu, Anmol, Thompson, Neil2025β€”1 articleView β†’
πŸ“„ PaperArxiv Goodhart RL StudyarXivβ˜…β˜…β˜…β˜†β˜†SummaryJacek Karwowski, Oliver Hayman, Xingjian Bai, Klaus Kiendlhofer, Charlie Griffin, Joar Skalse2023-10-13taxonomyreward-modeling+11 articleView β†’
πŸ“„ PaperArXiv training costsarXivβ˜…β˜…β˜…β˜†β˜†SummaryBen Cottier, Robi Rahman, Loredana Fattorini, Nestor Maslej, Tamay Besiroglu, David Owen2024-05-31trainingcompute+11 articleView β†’
πŸ“„ PaperarXiv, 2025arXivβ˜…β˜…β˜…β˜†β˜†SummarySeverin Field2025-01-25alignmentsafety+22 articlesView β†’
πŸ“„ PaperarXiv: Governance-as-a-Service - Multi-Agent Framework for AI CompliancearXivβ˜…β˜…β˜…β˜†β˜†SummarySuyash Gaurav, Jukka Heikkonen, Jatin Chaudhary2025-08-26alignmentgovernance+31 articleView β†’
πŸ“„ PaperArXiv: Thousands of AI Authors on the Future of AIarXivβ˜…β˜…β˜…β˜†β˜†SummaryKatja Grace, Harlan Stewart, Julia Fabienne SandkΓΌhler, Stephen Thomas, Ben Weinstein-Raun, Jan Brauner, Richard C. Korzekwa2024-01-05x-risktraining+21 articleView β†’
πŸ“„ PaperarXiv:2411.15114arXivβ˜…β˜…β˜…β˜†β˜†SummaryHjalmar Wijk, Tao Lin, Joel Becker, Sami Jawhar, Neev Parikh, Thomas Broadley, Lawrence Chan, Michael Chen, Josh Clymer, Jai Dhyani, Elena Ericheva, Katharyn Garcia, Brian Goodrich, Nikola Jurkovic, Holden Karnofsky, Megan Kinniment, Aron Lajko, Seraphina Nix, Lucas Sato, William Saunders, Maksym Taran, Ben West, Elizabeth Barnes2024-11-22capabilitiessafety+31 articleView β†’
πŸ“„ PaperarXiv:2503.14499arXivβ˜…β˜…β˜…β˜†β˜†SummaryThomas Kwa, Ben West, Joel Becker, Amy Deng, Katharyn Garcia, Max Hasin, Sami Jawhar, Megan Kinniment, Nate Rush, Sydney Von Arx, Ryan Bloom, Thomas Broadley, Haoxing Du, Brian Goodrich, Nikola Jurkovic, Luke Harold Miles, Seraphina Nix, Tao Lin, Neev Parikh, David Rein, Lucas Jun Koba Sato, Hjalmar Wijk, Daniel M. Ziegler, Elizabeth Barnes, Lawrence Chan2025-03-18capabilitiessafety+31 articleView β†’
πŸ“„ Paperas few as 200 fine-tuning examplesarXivβ˜…β˜…β˜…β˜†β˜†SummaryYeeun Kim, Hyunseo Shin, Eunkyung Choi, Hongseok Oh, Hyunjun Kim, Wonseok Hwang2024-03-11trainingopen-source+11 articleView β†’
πŸ”— WebAsilomar Conferenceβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“– ReferenceAsilomar precedentWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebASPI reportβ€”Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ›οΈ GovernmentASSESS AI ActUS Congressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAtlantic CouncilAtlantic Councilβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ”— WebAtlantic CouncilAtlantic Councilβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ”— WebAtlantic CouncilAtlantic Councilβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecoordination+11 articleView β†’
πŸ”— WebAtlantic Council DFRLabAtlantic Council

The Atlantic Council's DFRLab is a research organization focused on exposing digital threats, disinformation, and protecting democratic institutions through open-source investigations.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”open-sourceauthoritarianism+32 articlesView β†’
πŸ”— WebAtlantic Council: Digital TrustAtlantic Councilβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”institutional-trustsocial-capital+11 articleView β†’
✏️ Blog Postattempted game hacking 37%LessWrongβ˜…β˜…β˜…β˜†β˜†Metadatatechnicalities, Stag, Stephen McAleese, jordine, Dr. David Mathers2024-12-29cybersecurity1 articleView β†’
πŸ“„ PaperAttention Is All You NeedarXivβ˜…β˜…β˜…β˜†β˜†SummaryAshish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, Illia Polosukhin2017-06-12trainingcompute+11 articleView β†’
πŸ“„ PaperAttention mechanisms papersarXivβ˜…β˜…β˜…β˜†β˜†SummaryDzmitry Bahdanau, Kyunghyun Cho, Yoshua Bengio2014-09-01alignmentcapabilities+31 articleView β†’
πŸ”— WebAttestivβ€”Noneβ€”β€”social-engineeringvoice-cloning+11 articleView β†’
πŸ”— WebAttribution Patching: Activation Patching At Industrial Scaleβ€”Noneβ€”β€”sparse-autoencodersfeatures+11 articleView β†’
πŸ›οΈ GovernmentAuditing for Large Language ModelsCentre for the Governance of AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llm1 articleView β†’
πŸ“„ PaperAuditing language models for hidden objectivesAnthropic

Researchers deliberately created an AI model with a hidden reward model sycophancy objective and challenged blinded research teams to uncover its underlying motivations. The experiment aimed to develop and validate techniques for detecting concealed AI behaviors.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”llmpower-seeking+21 articleView β†’
πŸ”— WebAudrey Tangβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebAudrey Tangβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— Webaugmented realityMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebAugur Protocolβ€”Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebAugust 2024 "America in One Room: The Youth Vote"β€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebAugust 9, 2024, Illinois enacted HB 3773β€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentAum Shinrikyo's biological weapons programβ€”Noneβ€”β€”biosecurityprobability+21 articleView β†’
πŸ”— WebAustralia Groupβ€”Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ”— WebAustralian Strategic Policy Institute (ASPI)β€”Noneβ€”β€”governance1 articleView β†’
πŸ“„ PaperAuthentication systemsarXivβ˜…β˜…β˜…β˜†β˜†SummaryHuseyin Fuat Alsan, Taner Arsan2023-10-29capabilitiestraining+31 articleView β†’
πŸ”— WebAutomated interpretability agentβ€”Noneβ€”β€”interpretabilityeconomic1 articleView β†’
πŸ”— WebAutomation and Skill DecayGoogle Scholar

I apologize, but the source document appears to be a search results page with fragments of citations and abstracts, not a complete document. Without a coherent full text, I cannot comprehensively analyze this source as requested. The search results suggest multiple papers about skill decay and automation, but no single complete source is available. To properly complete the JSON template, I would need the full text of a specific research paper. If you'd like, I can: 1. Request the full text of a specific citation 2. Help you locate the complete source document 3. Provide a generalized analysis based on the citation fragments Would you like to proceed in one of those directions?

β˜…β˜…β˜…β˜…β˜†Summaryβ€”β€”economicautomation+21 articleView β†’
πŸ“– ReferenceAutomation biasWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”economicβ€”View β†’
πŸ“„ Paperautonomous vehicle planningarXivβ˜…β˜…β˜…β˜†β˜†SummaryArdi Tampuu, Maksym Semikin, Naveed Muhammad, Dmytro Fishman, Tambet Matiisen2020-03-13interpretabilitysafety+31 articleView β†’
πŸ”— WebAviation industry showsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAwesome Mechanistic Interpretability PapersGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”interpretabilitysparse-autoencoders+21 articleView β†’
πŸ”— WebAWSβ€”Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— WebAxios: Gemini 2.0 launch puts Google on road to AI agents (Dec 2024)β€”Noneβ€”β€”llm1 articleView β†’
πŸ”— WebAxios: Some AI dangers are already real, DeepMind's Hassabis says (Dec 2025)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAxios: Transformative AI is coming, and so are the risks (Dec 2025)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebAxis Intelligence

Comprehensive analysis of enterprise AI transformation reveals a systematic approach to achieving measurable business impact by 2025. The strategy focuses on organizational change, workflow redesign, and strategic implementation across multiple business functions.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebAxis Intelligence: AI Standards Guide 2025

The source provides an extensive overview of global AI standards, focusing on implementation strategies, regulatory requirements, and governance frameworks across industries. It offers practical guidance for organizations seeking to develop robust AI standards.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ”— WebAXRP Episode 34 - AI Evaluations with Beth Barnesβ€”Noneβ€”β€”evaluationevaluations+21 articleView β†’
πŸ”— WebBail et al. 2018PNASβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”epistemicnetwork-analysis+33 articlesView β†’
πŸ”— WebBank for International Settlementsβ€”Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ“„ PaperBansal et al. (2021)arXivβ˜…β˜…β˜…β˜†β˜†SummaryZana BuΓ§inca, Maja Barbara Malaya, Krzysztof Z. Gajos2021-02-19human-ai-interactionai-control+11 articleView β†’
πŸ”— WebBarbu et al. (2019)β€”Noneβ€”β€”robustnessgeneralization+11 articleView β†’
πŸ“„ PaperBaum (2017) - Survey of AI researchersarXivβ˜…β˜…β˜…β˜†β˜†SummaryKatja Grace, John Salvatier, Allan Dafoe, Baobao Zhang, Owain Evans2017-05-24governanceeconomic+31 articleView β†’
πŸ”— WebBBCβ€”Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— WebBBC interviewsβ€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebBBC: Deepfakes in Courtβ€”Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ“„ Paperbeat SAEs for interpretabilityarXivβ˜…β˜…β˜…β˜†β˜†SummaryGonΓ§alo Paulo, Stepan Shabalin, Nora Belrose2025-01-31interpretabilitysparse-autoencoders+21 articleView β†’
πŸ”— Webbecame the first stateβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperBeHonest: Benchmarking Honesty in Large Language ModelsarXivβ˜…β˜…β˜…β˜†β˜†SummarySteffi Chern, Zhulin Hu, Yuqing Yang, Ethan Chern, Yuan Guo, Jiahe Jin, Binjie Wang, Pengfei Liu2024-06-19alignmentcapabilities+3β€”View β†’
πŸ”— WebBelfer Centerβ€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ“„ PaperBelgian research (n = 1,579)β€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebBellingcat: Open source investigation

Bellingcat is a pioneering open-source investigation platform that uses digital forensics, geolocation, and AI to investigate complex global conflicts and technological issues.

β€”Fullβ€”β€”open-sourcehistorical-evidence+21 articleView β†’
πŸ”— WebBelt and Road Initiativeβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebBenchmarking AI Agents 2025

The document explores critical approaches to evaluating AI agent performance in 2025, highlighting key metrics, challenges, and emerging benchmarking tools and techniques.

β€”Fullβ€”β€”capabilitiesevaluationβ€”View β†’
πŸ”— WebBender, Gebru et al., 2021β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebBennett Institute, Cambridgeβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebBerg et al. (2008)ScienceDirect

A study comparing prediction markets to polls across five U.S. Presidential elections found that market predictions were closer to the eventual outcome 74% of the time, particularly when forecasting over 100 days in advance.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”β€”View β†’
πŸ”— WebBerg et al. (2008)β€”Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebBerkeley AIβ€”Noneβ€”β€”frameworkinstrumental-goals+11 articleView β†’
πŸ”— WebBerkeley AI Research: Detection methodsβ€”Noneβ€”β€”deepfakescontent-verification+32 articlesView β†’
πŸ”— WebBerkeley CHAI Researchβ€”Noneβ€”β€”networksrisk-interactions+11 articleView β†’
πŸ”— WebBerkeley CLTC Working Paper on Intolerable Risk Thresholdsβ€”Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebBerkeley's CLTC researchβ€”Noneβ€”β€”corrigibilityshutdown-problem+11 articleView β†’
πŸ”— WebBerkman Klein Center (Harvard)

Harvard's Berkman Klein Center conducts multidisciplinary research on AI's societal implications, focusing on ethics, governance, and legal challenges. The center brings together academics and practitioners to examine emerging technological landscapes.

β€”Fullβ€”β€”governancemental-health+21 articleView β†’
πŸ“„ PaperBesiroglu et al.arXivβ˜…β˜…β˜…β˜†β˜†SummaryEge Erdil, Tamay Besiroglu2022-12-10capabilitiescompute+31 articleView β†’
πŸ”— WebBest LLMs for Extended Context Windows

Research evaluated 22 AI models' ability to maintain context and retrieve information across long documents. Findings showed most models perform unreliably well before their claimed maximum context window.

β€”Fullβ€”β€”evaluationllmβ€”View β†’
πŸ”— WebBetfair Exchangeβ€”Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebBeth Barnesβ€”Noneβ€”β€”evaluationsdangerous-capabilities+11 articleView β†’
πŸ”— WebBeth Barnes - Safety evaluations and standards for AI (EA Forum)EA Forumβ˜…β˜…β˜…β˜†β˜†MetadataBeth Barnes2023-06-16safetyevaluation+31 articleView β†’
πŸ”— WebBetween 2009 and 2018Carnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ›οΈ GovernmentBiden Administration AI Executive Order 14110White Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”software-engineeringcode-generation+310 articlesView β†’
πŸ“– ReferenceBiden's EO 14110Wikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebBig Data & Society (2025)SAGE Journalsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alignmenttruthfulness+11 articleView β†’
πŸ”— WebBig Fund IIIβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebBig Tech controls 66% of cloud computingβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebBig Tech firms spent $61.5 million on lobbying in 2024β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebBig Tech's Cloud Oligopoly

A detailed analysis reveals how major tech companies like Microsoft, Amazon, and Google are dominating the AI and cloud computing markets through strategic investments and infrastructure control.

β€”Fullβ€”β€”x-riskirreversibility+12 articlesView β†’
πŸ“„ PaperBIG-Bench 2022arXivβ˜…β˜…β˜…β˜†β˜†SummaryAarohi Srivastava, Abhinav Rastogi, Abhishek Rao, Abu Awal Md Shoeb, Abubakar Abid, Adam Fisch, Adam R. Brown, Adam Santoro, Aditya Gupta, AdriΓ  Garriga-Alonso, Agnieszka Kluska, Aitor Lewkowycz, Akshat Agarwal, Alethea Power, Alex Ray, Alex Warstadt, Alexander W. Kocurek, Ali Safaya, Ali Tazarv, Alice Xiang, Alicia Parrish, Allen Nie, Aman Hussain, Amanda Askell, Amanda Dsouza, Ambrose Slone, Ameet Rahane, Anantharaman S. Iyer, Anders Andreassen, Andrea Madotto, Andrea Santilli, Andreas StuhlmΓΌller, Andrew Dai, Andrew La, Andrew Lampinen, Andy Zou, Angela Jiang, Angelica Chen, Anh Vuong, Animesh Gupta, Anna Gottardi, Antonio Norelli, Anu Venkatesh, Arash Gholamidavoodi, Arfa Tabassum, Arul Menezes, Arun Kirubarajan, Asher Mullokandov, Ashish Sabharwal, Austin Herrick, Avia Efrat, Aykut Erdem, Ayla Karakaş, B. Ryan Roberts, Bao Sheng Loe, Barret Zoph, BartΕ‚omiej Bojanowski, Batuhan Γ–zyurt, Behnam Hedayatnia, Behnam Neyshabur, Benjamin Inden, Benno Stein, Berk Ekmekci, Bill Yuchen Lin, Blake Howald, Bryan Orinion, Cameron Diao, Cameron Dour, Catherine Stinson, Cedrick Argueta, CΓ©sar Ferri RamΓ­rez, Chandan Singh, Charles Rathkopf, Chenlin Meng, Chitta Baral, Chiyu Wu, Chris Callison-Burch, Chris Waites, Christian Voigt, Christopher D. Manning, Christopher Potts, Cindy Ramirez, Clara E. Rivera, Clemencia Siro, Colin Raffel, Courtney Ashcraft, Cristina Garbacea, Damien Sileo, Dan Garrette, Dan Hendrycks, Dan Kilman, Dan Roth, Daniel Freeman, Daniel Khashabi, Daniel Levy, Daniel MoseguΓ­ GonzΓ‘lez, Danielle Perszyk, Danny Hernandez, Danqi Chen, Daphne Ippolito, Dar Gilboa, David Dohan, David Drakard, David Jurgens, Debajyoti Datta, Deep Ganguli, Denis Emelin, Denis Kleyko, Deniz Yuret, Derek Chen, Derek Tam, Dieuwke Hupkes, Diganta Misra, Dilyar Buzan, Dimitri Coelho Mollo, Diyi Yang, Dong-Ho Lee, Dylan Schrader, Ekaterina Shutova, Ekin Dogus Cubuk, Elad Segal, Eleanor Hagerman, Elizabeth Barnes, Elizabeth Donoway, Ellie Pavlick, Emanuele Rodola, Emma Lam, Eric Chu, Eric Tang, Erkut Erdem, Ernie Chang, Ethan A. Chi, Ethan Dyer, Ethan Jerzak, Ethan Kim, Eunice Engefu Manyasi, Evgenii Zheltonozhskii, Fanyue Xia, Fatemeh Siar, Fernando MartΓ­nez-Plumed, Francesca HappΓ©, Francois Chollet, Frieda Rong, Gaurav Mishra, Genta Indra Winata, Gerard de Melo, GermΓ‘n Kruszewski, Giambattista Parascandolo, Giorgio Mariani, Gloria Wang, Gonzalo Jaimovitch-LΓ³pez, Gregor Betz, Guy Gur-Ari, Hana Galijasevic, Hannah Kim, Hannah Rashkin, Hannaneh Hajishirzi, Harsh Mehta, Hayden Bogar, Henry Shevlin, Hinrich SchΓΌtze, Hiromu Yakura, Hongming Zhang, Hugh Mee Wong, Ian Ng, Isaac Noble, Jaap Jumelet, Jack Geissinger, Jackson Kernion, Jacob Hilton, Jaehoon Lee, Jaime FernΓ‘ndez Fisac, James B. Simon, James Koppel, James Zheng, James Zou, Jan KocoΕ„, Jana Thompson, Janelle Wingfield, Jared Kaplan, Jarema Radom, Jascha Sohl-Dickstein, Jason Phang, Jason Wei, Jason Yosinski, Jekaterina Novikova, Jelle Bosscher, Jennifer Marsh, Jeremy Kim, Jeroen Taal, Jesse Engel, Jesujoba Alabi, Jiacheng Xu, Jiaming Song, Jillian Tang, Joan Waweru, John Burden, John Miller, John U. Balis, Jonathan Batchelder, Jonathan Berant, JΓΆrg Frohberg, Jos Rozen, Jose Hernandez-Orallo, Joseph Boudeman, Joseph Guerr, Joseph Jones, Joshua B. Tenenbaum, Joshua S. Rule, Joyce Chua, Kamil Kanclerz, Karen Livescu, Karl Krauth, Karthik Gopalakrishnan, Katerina Ignatyeva, Katja Markert, Kaustubh D. Dhole, Kevin Gimpel, Kevin Omondi, Kory Mathewson, Kristen Chiafullo, Ksenia Shkaruta, Kumar Shridhar, Kyle McDonell, Kyle Richardson, Laria Reynolds, Leo Gao, Li Zhang, Liam Dugan, Lianhui Qin, Lidia Contreras-Ochando, Louis-Philippe Morency, Luca Moschella, Lucas Lam, Lucy Noble, Ludwig Schmidt, Luheng He, Luis Oliveros ColΓ³n, Luke Metz, LΓΌtfi Kerem Şenel, Maarten Bosma, Maarten Sap, Maartje ter Hoeve, Maheen Farooqi, Manaal Faruqui, Mantas Mazeika, Marco Baturan, Marco Marelli, Marco Maru, Maria Jose RamΓ­rez Quintana, Marie Tolkiehn, Mario Giulianelli, Martha Lewis, Martin Potthast, Matthew L. Leavitt, Matthias Hagen, MΓ‘tyΓ‘s Schubert, Medina Orduna Baitemirova, Melody Arnaud, Melvin McElrath, Michael A. Yee, Michael Cohen, Michael Gu, Michael Ivanitskiy, Michael Starritt, Michael Strube, MichaΕ‚ SwΔ™drowski, Michele Bevilacqua, Michihiro Yasunaga, Mihir Kale, Mike Cain, Mimee Xu, Mirac Suzgun, Mitch Walker, Mo Tiwari, Mohit Bansal, Moin Aminnaseri, Mor Geva, Mozhdeh Gheini, Mukund Varma T, Nanyun Peng, Nathan A. Chi, Nayeon Lee, Neta Gur-Ari Krakover, Nicholas Cameron, Nicholas Roberts, Nick Doiron, Nicole Martinez, Nikita Nangia, Niklas Deckers, Niklas Muennighoff, Nitish Shirish Keskar, Niveditha S. Iyer, Noah Constant, Noah Fiedel, Nuan Wen, Oliver Zhang, Omar Agha, Omar Elbaghdadi, Omer Levy, Owain Evans, Pablo Antonio Moreno Casares, Parth Doshi, Pascale Fung, Paul Pu Liang, Paul Vicol, Pegah Alipoormolabashi, Peiyuan Liao, Percy Liang, Peter Chang, Peter Eckersley, Phu Mon Htut, Pinyu Hwang, Piotr MiΕ‚kowski, Piyush Patil, Pouya Pezeshkpour, Priti Oli, Qiaozhu Mei, Qing Lyu, Qinlang Chen, Rabin Banjade, Rachel Etta Rudolph, Raefer Gabriel, Rahel Habacker, Ramon Risco, RaphaΓ«l MilliΓ¨re, Rhythm Garg, Richard Barnes, Rif A. Saurous, Riku Arakawa, Robbe Raymaekers, Robert Frank, Rohan Sikand, Roman Novak, Roman Sitelew, Ronan LeBras, Rosanne Liu, Rowan Jacobs, Rui Zhang, Ruslan Salakhutdinov, Ryan Chi, Ryan Lee, Ryan Stovall, Ryan Teehan, Rylan Yang, Sahib Singh, Saif M. Mohammad, Sajant Anand, Sam Dillavou, Sam Shleifer, Sam Wiseman, Samuel Gruetter, Samuel R. Bowman, Samuel S. Schoenholz, Sanghyun Han, Sanjeev Kwatra, Sarah A. Rous, Sarik Ghazarian, Sayan Ghosh, Sean Casey, Sebastian Bischoff, Sebastian Gehrmann, Sebastian Schuster, Sepideh Sadeghi, Shadi Hamdan, Sharon Zhou, Shashank Srivastava, Sherry Shi, Shikhar Singh, Shima Asaadi, Shixiang Shane Gu, Shubh Pachchigar, Shubham Toshniwal, Shyam Upadhyay, Shyamolima, Debnath, Siamak Shakeri, Simon Thormeyer, Simone Melzi, Siva Reddy, Sneha Priscilla Makini, Soo-Hwan Lee, Spencer Torene, Sriharsha Hatwar, Stanislas Dehaene, Stefan Divic, Stefano Ermon, Stella Biderman, Stephanie Lin, Stephen Prasad, Steven T. Piantadosi, Stuart M. Shieber, Summer Misherghi, Svetlana Kiritchenko, Swaroop Mishra, Tal Linzen, Tal Schuster, Tao Li, Tao Yu, Tariq Ali, Tatsu Hashimoto, Te-Lin Wu, ThΓ©o Desbordes, Theodore Rothschild, Thomas Phan, Tianle Wang, Tiberius Nkinyili, Timo Schick, Timofei Kornev, Titus Tunduny, Tobias Gerstenberg, Trenton Chang, Trishala Neeraj, Tushar Khot, Tyler Shultz, Uri Shaham, Vedant Misra, Vera Demberg, Victoria Nyamai, Vikas Raunak, Vinay Ramasesh, Vinay Uday Prabhu, Vishakh Padmakumar, Vivek Srikumar, William Fedus, William Saunders, William Zhang, Wout Vossen, Xiang Ren, Xiaoyu Tong, Xinran Zhao, Xinyi Wu, Xudong Shen, Yadollah Yaghoobzadeh, Yair Lakretz, Yangqiu Song, Yasaman Bahri, Yejin Choi, Yichi Yang, Yiding Hao, Yifu Chen, Yonatan Belinkov, Yu Hou, Yufang Hou, Yuntao Bai, Zachary Seid, Zhuoye Zhao, Zijian Wang, Zijie J. Wang, Zirui Wang, Ziyi Wu2022-06-09capabilitiesevaluation+31 articleView β†’
πŸ”— WebBIG-Bench evaluation suiteGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”evaluationscaling+21 articleView β†’
πŸ”— WebBik et al.β€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebBik et al. (2016)β€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebBio anchors frameworkβ€”Noneβ€”β€”effective-altruismai-safety-funding+12 articlesView β†’
πŸ”— WebBio Anchors ReportOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”effective-altruismai-safety-funding+11 articleView β†’
πŸ”— WebBiological Weapons Conventionβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebBiometric Information Privacy Actβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebBiometric Updateβ€”Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ”— WebBiopharmaTrend report from April 2024β€”Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ“– ReferenceBiopreparatWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebBiopreparat programβ€”Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ”— WebBiosecurity resourcesβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebBipartisan Policy Center notesβ€”Noneβ€”β€”governance1 articleView β†’
πŸ”— WebBlackSuit ransomware group attacked CDK Globalβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebBletchley DeclarationBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebBletchley Declarationβ€”Noneβ€”β€”β€”β€”View β†’
πŸ›οΈ GovernmentBletchley DeclarationUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+13 articlesView β†’
πŸ“„ PaperBlitz: "Deepfakes and Evidence Law"SSRNβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ”— WebBloom: Automated Behavioral EvaluationsAnthropic Alignmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationeconomic+31 articleView β†’
πŸ”— WebBloombergβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebBloomberg Intelligence projectsβ€”Noneβ€”β€”frontier-labssafety-culture+11 articleView β†’
πŸ”— WebBloomberg, 2024β€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentBLS Employment Projections

The Bureau of Labor Statistics examines how AI might affect employment in different sectors, finding that productivity gains will vary by occupation but are unlikely to cause widespread job losses in the near term.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ›οΈ GovernmentBLS Employment Projections 2024-2034

The Bureau of Labor Statistics forecasts moderate employment growth of 3.1% from 2024-2034, with healthcare and technology sectors experiencing the most significant job increases.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ›οΈ GovernmentBLS Industry Projections

The Bureau of Labor Statistics forecasts total employment will grow to 174.6 million by 2033, with significant job gains in healthcare, professional services, and emerging technologies like clean energy and AI.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ›οΈ GovernmentBLS Labor Force Projections

The Bureau of Labor Statistics forecasts a continued slowdown in labor force and population growth through 2033, primarily due to an aging population and declining fertility rates. These trends will impact GDP growth, employment, and overall economic dynamics.

β€”Fullβ€”β€”economicagi1 articleView β†’
πŸ›οΈ GovernmentBLS Projections

Data scientist employment is expected to grow 34% from 2024-2034, with a median annual wage of $112,590. The field requires strong analytical and technical skills.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebBlueDot 2022 Cohort Analysisβ€”Noneβ€”β€”field-buildingtraining-programs+11 articleView β†’
πŸ”— WebBlueDot Impactβ€”Noneβ€”β€”field-buildingtraining-programs+11 articleView β†’
πŸ”— WebBlueprint Biosecurityβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebBook

The Oxford Internet Institute conducts interdisciplinary research on digital technologies' social and political implications, focusing on misinformation, computational propaganda, and platform governance.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ”— WebBook

The Shallows examines the cognitive impact of digital technology, arguing that internet use is rewiring our brains and reducing our capacity for deep, contemplative thought.

β€”Fullβ€”β€”automationhuman-factors+11 articleView β†’
πŸ”— WebBookβ€”Noneβ€”β€”economic-inequalitymarket-concentration+11 articleView β†’
πŸ”— WebBookβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebBookβ€”Noneβ€”β€”ai-ethicspersuasion+11 articleView β†’
πŸ”— WebBostrom (2014)β€”Noneβ€”β€”frameworkinstrumental-goals+11 articleView β†’
πŸ”— WebBostrom (2014): SuperintelligenceAmazonβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”agiintelligence-explosion+32 articlesView β†’
πŸ”— WebBostrom: Racing to the Precipiceβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperBounded objectives researcharXivβ˜…β˜…β˜…β˜†β˜†SummaryStuart Armstrong, SΓΆren Mindermann2017-12-15governancecausal-model+22 articlesView β†’
πŸ”— Webbrain-computer interfacesβ€”Noneβ€”β€”computedisinformation+21 articleView β†’
πŸ”— WebBrand protection firm MarkMonitor's analysisβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebBreaking Defenseβ€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebBrightside AIβ€”Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ”— WebBritish lawmakers accuse Google of 'breach of trust' over delayed Gemini 2.5 Pro safety reportFortune

A group of 60 U.K. lawmakers criticized Google DeepMind for not fully disclosing safety information about its Gemini 2.5 Pro AI model as previously committed. The letter argues the company failed to provide comprehensive model testing details.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”safetyevaluation+11 articleView β†’
πŸ”— WebBrookingsBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”social-engineeringmanipulation+11 articleView β†’
πŸ”— WebBrookingsBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebBrookingsBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic-inequalitymarket-concentration+11 articleView β†’
πŸ”— WebBrookingsBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebBrookings AI GovernanceBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancemarket-concentration+11 articleView β†’
πŸ”— WebBrookings AI GovernanceBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancehuman-agency+21 articleView β†’
πŸ”— WebBrookings AI governance trackerBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governanceinterventions+33 articlesView β†’
πŸ”— WebBrookings Center for Technology InnovationBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebBrookings emphasizesBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebBrookings InstitutionBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— WebBrookings InstitutionBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— WebBrookings Institution researchBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ”— WebBrookings on NetworksBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebBrookings researchBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ”— WebBrookings researchers call the "liar's dividend"Brookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationdeepfakes+12 articlesView β†’
πŸ”— WebBrookings: AI CompetitionBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theoryinternational-coordination+32 articlesView β†’
πŸ”— WebBrookings: Misrepresentations of California's AI safety billBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyregulation+21 articleView β†’
πŸ”— WebBrookings: Trust in GovernmentBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”institutional-trustsocial-capital+11 articleView β†’
πŸ”— WebBrookings: Winner-Take-Most AI EconomyBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— Webbrookings.eduBrookings Institution

Brookings Institution provides commentary on AI policy, international cooperation, and global economic development. Explores potential challenges and implications of technological and geopolitical shifts.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governanceeconomicβ€”View β†’
πŸ”— Webbrookings.edu/research/governance-aiBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancehistorical-evidence+21 articleView β†’
πŸ“„ PaperBrown et al. (2020)arXivβ˜…β˜…β˜…β˜†β˜†SummaryTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, Mark Chen, Eric Sigler, Mateusz Litwin, Scott Gray, Benjamin Chess, Jack Clark, Christopher Berner, Sam McCandlish, Alec Radford, Ilya Sutskever, Dario Amodei2020-05-28capabilitiestraining+34 articlesView β†’
πŸ“„ PaperBrundage et al. (2024). "The Malicious Use of AI in Cybersecurity"arXivβ˜…β˜…β˜…β˜†β˜†SummaryShuai Li, Ming Gong, Yu-Hang Li, Hua Jiang, X. C. Xie2024-04-18cybersecuritytimeline+11 articleView β†’
πŸ”— WebBrynjolfsson & Mitchell (2017)β€”Noneβ€”β€”economic-inequalitymarket-concentration+11 articleView β†’
πŸ”— WebBuilding an early warning system for LLM-aided biological threat creationOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”biosecurityllm+21 articleView β†’
πŸ“„ PaperBuilding Machines That Learn and Think Like PeoplearXivβ˜…β˜…β˜…β˜†β˜†SummaryBrenden M. Lake, Tomer D. Ullman, Joshua B. Tenenbaum, Samuel J. Gershman2016-04-01capabilitiesbiosecurity1 articleView β†’
πŸ”— WebBulletin of Atomic Scientists: AI Surveillance and Democracyβ€”Noneβ€”β€”x-riskgovernance+12 articlesView β†’
πŸ”— WebBulletin of the Atomic Scientists arguesβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ›οΈ GovernmentBureau of Industry and SecurityBureau of Industry and Securityβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityauthoritarianism+21 articleView β†’
πŸ”— WebBureau of Industry and Security assessedβ€”Noneβ€”β€”cybersecurity1 articleView β†’
πŸ›οΈ GovernmentBureau of Labor Statisticsβ€”Noneβ€”β€”economicmonitoring+21 articleView β†’
πŸ“– ReferenceBWCWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebBWCUnited Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperByrne & Christopher (2020)Natureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ“„ PaperByrne & Christopher, 2020Natureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”β€”View β†’
πŸ”— WebC2PA Explainer Videos

The Coalition for Content Provenance and Authenticity (C2PA) offers a technical standard that acts like a 'nutrition label' for digital content, tracking its origin and edit history.

β€”Fullβ€”β€”epistemictimeline+311 articlesView β†’
πŸ”— WebC2PA Technical Specification

The C2PA Technical Specification provides a standardized framework for tracking and verifying the origin, modifications, and authenticity of digital content using cryptographic signatures and assertions.

β€”Fullβ€”β€”deepfakescontent-verification+32 articlesView β†’
πŸ“„ PaperCabanac et al. (2022)arXivβ˜…β˜…β˜…β˜†β˜†SummaryGuangyin Jin, Fuxian Li, Jinlei Zhang, Mudan Wang, Jincai Huang2022-07-22capabilitieseconomic+31 articleView β†’
πŸ“„ PaperCabanac et al., 2022arXivβ˜…β˜…β˜…β˜†β˜†SummaryGuillaume Cabanac, Cyril LabbΓ©, Alexander Magazinov2021-07-12interpretabilitydeception+1β€”View β†’
πŸ”— WebCAC warningβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCAIDPMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”self-regulationindustry-commitments+11 articleView β†’
πŸ”— WebCAIDP: International AI Treaty

The Council of Europe AI Treaty is a groundbreaking international convention aimed at ensuring AI systems respect human rights, democratic principles, and legal standards. It provides a comprehensive legal framework for AI development, use, and oversight across public and private sectors.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebCAIS 2024 Impact ReportCenter for AI Safetyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”field-buildingtraining-programs+11 articleView β†’
πŸ”— WebCAIS BlogCenter for AI Safetyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-safetyx-risk+11 articleView β†’
πŸ”— WebCAIS PublicationsCenter for AI Safetyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-safetyx-risk+11 articleView β†’
πŸ”— WebCAIS Survey Analysis

A Rethink Priorities survey analyzed responses from people disagreeing with the CAIS statement about AI extinction risk. Key themes included prioritizing other issues and skepticism about AI's potential for causing extinction.

β€”Fullβ€”β€”x-riskβ€”View β†’
πŸ”— WebCAIS SurveysCenter for AI Safety

The Center for AI Safety conducts technical and conceptual research to mitigate potential catastrophic risks from advanced AI systems. They take a comprehensive approach spanning technical research, philosophy, and societal implications.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”safetyx-risk+323 articlesView β†’
πŸ›οΈ GovernmentCAISI Evaluation of DeepSeek AI Models Finds Shortcomings and RisksNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”evaluationgame-theory+21 articleView β†’
πŸ›οΈ GovernmentCalifornia Assembly Privacy and Consumer Protection Committee Analysisβ€”Noneβ€”β€”regulationstate-policy+11 articleView β†’
πŸ›οΈ GovernmentCalifornia Consumer Privacy Actβ€”Metadataβ€”2018ai-ethicspersuasion+11 articleView β†’
πŸ›οΈ GovernmentCalifornia SB 1001β€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ›οΈ GovernmentCalifornia's AB 2655β€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebCalled "meager" and "worrisome"Fortuneβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webcalled for explicit US-China collaborationβ€”Noneβ€”β€”game-theorycoordination+12 articlesView β†’
πŸ”— Webcalled the Paris Summit a "missed opportunity"β€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebCalMatters: Newsom vetoes major California artificial intelligence billβ€”Noneβ€”β€”regulationstate-policy+11 articleView β†’
πŸ”— WebCambridgeβ€”Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ”— WebCambridgeβ€”Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ”— WebCambridge AI Safetyβ€”Noneβ€”β€”safetysocial-engineering+21 articleView β†’
πŸ“„ PaperCambridge Analytica case studyNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebCambridge Analytica revelationsβ€”Noneβ€”β€”ai-ethicspersuasion+11 articleView β†’
πŸ“„ PaperCambridge Analytica-styleNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebCambridge Data & Policy StudyCambridge University Pressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”governance1 articleView β†’
πŸ”— WebCan Preparedness Frameworks Pull Their Weight?β€”Noneβ€”β€”benchmarksred-teaming+12 articlesView β†’
πŸ”— Webcan worsen with model sizeβ€”Noneβ€”β€”traininghuman-feedback+11 articleView β†’
πŸ”— WebCanadian Parliament 2023β€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ“„ PaperCapability Control MethodsarXivβ˜…β˜…β˜…β˜†β˜†SummaryRonald Cardenas, Bingsheng Yao, Dakuo Wang, Yufang Hou2023-10-23capabilitieseconomic+31 articleView β†’
πŸ”— WebCarl Shulman and colleagues80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ“„ PaperCarlini et al. (2023)arXivβ˜…β˜…β˜…β˜†β˜†SummaryA. Ismael, S. Khalil2023-01-05β€”1 articleView β†’
πŸ”— WebCarlisle, 2017β€”Noneβ€”β€”β€”β€”View β†’
πŸ“„ PaperCarlsmith (2021)arXivβ˜…β˜…β˜…β˜†β˜†SummaryYixuan Su, David Vandyke, Sihui Wang, Yimai Fang, Nigel Collier2021-08-31capabilitiesevaluation+31 articleView β†’
πŸ“„ PaperCarlsmith (2021)arXivβ˜…β˜…β˜…β˜†β˜†SummaryV. Yu. Irkhin, Yu. N. Skryabin2021-08-31formal-analysispower-seeking+11 articleView β†’
πŸ“„ PaperCarlsmith (2023) - Scheming AIsarXivβ˜…β˜…β˜…β˜†β˜†SummaryJoe Carlsmith2023-11-14alignmentcapabilities+32 articlesView β†’
πŸ“„ PaperCarlsmith (2024): AI Forecasting for Existential RiskarXivβ˜…β˜…β˜…β˜†β˜†SummaryElliot J. Carr2024-01-17interpretabilityx-risk+31 articleView β†’
πŸ”— WebCarnegie analysis warnsCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theorycoordination+12 articlesView β†’
πŸ”— WebCarnegie EndowmentCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCarnegie Endowment - Can Democracy Survive the Disruptive Power of AI?Carnegie Endowment

The article explores how advanced AI technologies can destabilize democratic systems by enabling rapid creation of synthetic content and foreign interference. It examines the risks of AI-generated misinformation and proposes multi-stakeholder strategies to mitigate these challenges.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”democratic-innovationcollective-intelligence+12 articlesView β†’
πŸ”— WebCarnegie Endowment AI Global Surveillance IndexCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”authoritarianismstability+11 articleView β†’
πŸ”— WebCarnegie Endowment analysisCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCarnegie Endowment for International PeaceCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskgovernance+32 articlesView β†’
πŸ”— WebCarnegie Endowment researchCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”regulationchina+11 articleView β†’
πŸ”— WebCarnegie Endowment: AI Governance Arms RaceCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governanceβ€”View β†’
πŸ”— WebCarnegie Endowment: All Eyes on Sacramento: SB 1047 and the AI Safety DebateCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyregulation+21 articleView β†’
πŸ”— WebCarnegie Endowment: If-Then Commitments for AI Risk ReductionCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCarnegie Endowment: The AI Governance Arms RaceCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governance1 articleView β†’
πŸ“„ PaperCasper, S., et al. (2024). "Black-Box Access is Insufficient for Rigorous AI Audits."arXivβ˜…β˜…β˜…β˜†β˜†SummaryStephen Casper, Carson Ezell, Charlotte Siegmann, Noam Kolt, Taylor Lynn Curtis, Benjamin Bucknall, Andreas Haupt, Kevin Wei, JΓ©rΓ©my Scheurer, Marius Hobbhahn, Lee Sharkey, Satyapriya Krishna, Marvin Von Hagen, Silas Alberti, Alan Chan, Qinyi Sun, Michael Gerovitch, David Bau, Max Tegmark, David Krueger, Dylan Hadfield-Menell2024-01-25governancesafety+31 articleView β†’
✏️ Blog PostCatalyze's pilot programEA Forumβ˜…β˜…β˜…β˜†β˜†MetadataCatalyze Impact, Alexandra Bos, Mick2025-09-16field-buildingtraining-programs+11 articleView β†’
πŸ“„ PaperCausal Representation Learning for AI SafetyarXivβ˜…β˜…β˜…β˜†β˜†SummaryThomas Krendl Gilbert, Sarah Dean, Tom Zick, Nathan Lambert2022-02-11governancesafety+31 articleView β†’
πŸ”— WebCausal Scrubbingβ€”Noneβ€”β€”mesa-optimizationinner-alignment+33 articlesView β†’
πŸ“„ PaperCausal Scrubbing PaperarXivβ˜…β˜…β˜…β˜†β˜†SummaryJing Huang, Atticus Geiger, Karel D'Oosterlinck, Zhengxuan Wu, Christopher Potts2023-09-19evaluationllm+31 articleView β†’
πŸ”— Webcauses of sycophantic behaviorβ€”Noneβ€”β€”alignmenttruthfulness+11 articleView β†’
πŸ”— WebCB Insightsβ€”Noneβ€”β€”ai-forecastingcompute-trends+11 articleView β†’
πŸ”— WebCBS 60 Minutesβ€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ“– ReferenceCCDHWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ›οΈ GovernmentCDCβ€”Noneβ€”β€”monitoringearly-warning+11 articleView β†’
πŸ›οΈ GovernmentCDC Select Agentsβ€”Noneβ€”β€”monitoringearly-warning+11 articleView β†’
πŸ›οΈ Governmentcdc.gov/biowatchβ€”Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ›οΈ GovernmentCDC's BioWatch programβ€”Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ”— WebCDK reportedly paid \$25 million in bitcoinβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebCEN-CENELEC JTC 21β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCenter for AI SafetyEA Forum

The Center for AI Safety conducts technical and conceptual research on AI safety, advocates for responsible AI development, and supports the AI safety research community through various initiatives.

β˜…β˜…β˜…β˜†β˜†FullCenter for AI Safety, Corin Katzke, Dan H2024-12-19safetyβ€”View β†’
✏️ Blog PostCenter for AI Safety 2024 Year in Review (EA Forum)EA Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebCenter for AI Safety SafeBench competitionβ€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebCenter for AI Safety: Catastrophic RisksCenter for AI Safetyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyx-riskβ€”View β†’
πŸ”— WebCenter for Human-Compatible AI

The Center for Human-Compatible AI (CHAI) focuses on reorienting AI research towards developing systems that are fundamentally beneficial and aligned with human values through technical and conceptual innovations.

β€”Fullβ€”β€”alignmentagentic+312 articlesView β†’
πŸ”— WebCenter for Human-Compatible AIβ€”Noneβ€”β€”inner-alignmentdistribution-shift+11 articleView β†’
πŸ”— WebCenter for Humane Technologyβ€”MetadataCenter for Humane Technology, Substackβ€”mental-healthai-ethics+32 articlesView β†’
πŸ”— WebCenter for Humane Technologyβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebCenter for Security and Emerging TechnologyCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityinterventions+21 articleView β†’
πŸ”— WebCenter for Security and Emerging Technology analysisCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityopen-source+21 articleView β†’
πŸ”— WebCenter for Strategic and International StudiesCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationtiming+31 articleView β†’
πŸ”— WebCenter for Strategic and International StudiesCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebCenter for Strategic and International StudiesCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecoordination+11 articleView β†’
πŸ”— WebCenter for Strategic StudiesCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”escalationconflict+33 articlesView β†’
πŸ”— WebCentre for Effective Altruismβ€”Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ”— WebCentre for Future Generations: The AI Safety Institute Networkβ€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebCentre for Long-Term Resilience: AI Safety Frameworks Risk Governanceβ€”Noneβ€”β€”governancesafety1 articleView β†’
πŸ”— WebCEPA - AI and Arms Races

The article critiques the 'AI arms race' concept, arguing that AI competition is fundamentally different from traditional arms races and requires a more nuanced understanding of technological development.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebCEPR analysisβ€”Noneβ€”β€”x-riskvalue-lock-in+11 articleView β†’
πŸ”— WebCESI OECD Analysis

The OECD's 2023 Employment Outlook highlights significant job risks from AI, with 27% of jobs potentially automatable and workers expressing concerns about job displacement.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ›οΈ GovernmentCFIUS review processβ€”Noneβ€”β€”risk-factordiffusion+11 articleView β†’
πŸ”— WebCFRβ€”Noneβ€”β€”social-engineeringmanipulation+11 articleView β†’
πŸ”— Webcfr.org/backgrounder/artificial-intelligence-and-national-securityβ€”Noneβ€”β€”cybersecurityhistorical-evidence+21 articleView β†’
πŸ›οΈ GovernmentCFTCβ€”Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebCHAI Newsβ€”Noneβ€”β€”inverse-reinforcement-learningvalue-learning+11 articleView β†’
πŸ”— WebCHAI Papersβ€”Noneβ€”β€”inverse-reinforcement-learningvalue-learning+11 articleView β†’
πŸ”— WebCHAI Teamβ€”Noneβ€”β€”inverse-reinforcement-learningvalue-learning+11 articleView β†’
πŸ“„ PaperChain-of-thought analysisarXivβ˜…β˜…β˜…β˜†β˜†SummaryJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, Denny Zhou2022-01-28capabilitiesevaluation+34 articlesView β†’
πŸ”— WebChallenges in automating fact-checkingSAGE Journalsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economicknowledge-management+21 articleView β†’
πŸ”— WebChatGPT 4.5 Jailbreaking & Red Teaming Analysis

A comprehensive security audit of ChatGPT 4.5 demonstrates strong resistance to jailbreaking attempts, with 97% of bypass attempts blocked and a 99% overall safe response rate.

β€”Fullβ€”β€”safetycybersecurityβ€”View β†’
πŸ”— WebChatGPT launchOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorgame-theory+32 articlesView β†’
πŸ”— WebChatGPT's November 2022 launchOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecoordination+11 articleView β†’
πŸ”— WebChatham Houseβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperChemBencharXivβ˜…β˜…β˜…β˜†β˜†SummaryLei Yao, Yong Zhang, Zilong Yan, Jialu Tian2023-10-13capabilitiesllm+31 articleView β†’
πŸ”— WebChen & Plott (2002)β€”Noneβ€”β€”β€”β€”View β†’
πŸ“„ PaperChen & Plott (2002)Springerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebChesney & Citron (2019)β€”Noneβ€”β€”deepfakesdigital-evidence+12 articlesView β†’
πŸ“„ PaperChesney & Citron: "Deep Fakes and the Infocalypse"SSRNβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ”— WebChina AI Regulatory Trackerβ€”Noneβ€”β€”governanceregulation+21 articleView β†’
πŸ”— WebChina-US dialogue progressβ€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebChina's AI Chip Deficit: Why Huawei Can't Catch Nvidiaβ€”Noneβ€”β€”compute4 articlesView β†’
πŸ”— WebChina's DeepSeek R1β€”Noneβ€”β€”governancecoordination+11 articleView β†’
πŸ›οΈ GovernmentChina's Global AI Governance Action Planβ€”Noneβ€”β€”governance2 articlesView β†’
πŸ”— WebChina's military AI developmentCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— WebChina's Quest for Semiconductor Self-Sufficiencyβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebChinese-backed influence campaignβ€”Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ›οΈ GovernmentCHIPS ActWhite Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”computeprioritization+21 articleView β†’
πŸ›οΈ GovernmentCHIPS and Science ActNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”computegovernance+32 articlesView β†’
πŸ”— WebChris Olah80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”ai-safetyconstitutional-ai+11 articleView β†’
πŸ”— WebChristiano (2018)β€”Noneβ€”β€”cost-effectivenessresearch-priorities+12 articlesView β†’
πŸ”— WebChristiano, P. (2017). "Corrigibility."β€”Noneβ€”β€”shutdown-problemai-control+11 articleView β†’
πŸ”— WebCIFARβ€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebCIGI: The Silent Erosionβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebCimplifi: Updated State of AI Regulations for 2025

Comprehensive overview of AI regulatory developments in 2024-2025, highlighting emerging national and regional approaches to AI governance and legislation.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ”— Webcircuit tracing researchTransformer Circuitsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”sparse-autoencodersfeatures+11 articleView β†’
πŸ”— WebCircuits UpdatesTransformer Circuitsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interpretabilityai-safety+11 articleView β†’
πŸ”— WebCircuits Updates - July 2025Transformer Circuitsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interpretabilitysparse-autoencoders+21 articleView β†’
πŸ”— WebCircuits workβ€”Noneβ€”β€”interpretability2 articlesView β†’
πŸ”— WebCIRL corrigibility proved fragileMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”power-seekingself-preservation+11 articleView β†’
πŸ›οΈ GovernmentCISACISAβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”probabilitydecomposition+34 articlesView β†’
πŸ›οΈ GovernmentCISACISAβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ›οΈ GovernmentCISA Cybersecurity VideosCISAβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ›οΈ GovernmentCISA OT AI integration principlesCISAβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ›οΈ GovernmentCISA Roadmap for AICISAβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebCitizen Lab - Digital Rights Researchβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebCitizens' assemblies on AIβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebClaimReview schema

ClaimReview is a Schema.org type for systematically documenting claim reviews, including the claim, reviewer, rating, and context of the original statement.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebClaudeAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llmprobability+33 articlesView β†’
πŸ”— WebClaude 3.7 SonnetAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llm2 articlesView β†’
πŸ”— WebClaude with computer useAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”computellm+33 articlesView β†’
πŸ”— WebClaude's constitutionAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llmai-safety+33 articlesView β†’
πŸ”— Webclear disclosure when AI contributes to consequential decisionsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webcloud AI servicesβ€”Noneβ€”β€”formal-analysispower-seeking+11 articleView β†’
πŸ”— WebCloud laboratoriesRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economicbiosecurity+21 articleView β†’
πŸ”— WebCMU thesis (2024)β€”Noneβ€”β€”robustnessgeneralization+11 articleView β†’
πŸ”— WebCnAISDAβ€”Noneβ€”β€”regulationchina+13 articlesView β†’
πŸ”— WebCnAISDA establishmentCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”regulationchina+11 articleView β†’
πŸ”— WebCnAISDA launched February 2025Carnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”regulationchina+11 articleView β†’
πŸ”— WebCNASCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”agenticplanning+314 articlesView β†’
πŸ”— WebCNASCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”social-engineeringmanipulation+11 articleView β†’
πŸ”— WebCNASCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationresource-allocation+11 articleView β†’
πŸ”— WebCNASCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorcompetition+11 articleView β†’
πŸ”— WebCNASCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCNASCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCNAS AI governance surveyCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governanceregulation+21 articleView β†’
πŸ”— WebCNAS AI PolicyCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancenetworks+33 articlesView β†’
πŸ”— WebCNAS analysisCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCNAS analysisCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebCNAS reportCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”biosecuritydual-use-research+12 articlesView β†’
πŸ”— WebCNAS researchCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebCNAS researchCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCNAS: Technology CompetitionCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— WebCNAS's "Secure, Governable Chips" reportCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”compute1 articleView β†’
πŸ”— WebCNBCCNBC

A group of seven tech startups tracked by Forge Global has nearly doubled in value to $1.3 trillion, with AI companies leading the surge. OpenAI, Anthropic, and xAI are at the forefront of this explosive growth.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebCNBC AnthropicCNBC

Microsoft and Nvidia are making substantial investments in Anthropic, expanding their AI partnerships and computing capacity. The deal positions Anthropic as a major player in the AI landscape.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebCNBC Nvidia market analysisCNBC

Nvidia controls the majority of the AI chip market, with unprecedented market capitalization and revenue driven by AI accelerator demand. Competitors are emerging from tech giants, startups, and chipmakers seeking to challenge Nvidia's dominance.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”compute1 articleView β†’
πŸ”— WebCNBC: Inside Isomorphic Labs (Apr 2025)CNBCβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCO/AI analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCoalition for Networked Information

CNI is a collaborative organization advancing information technology in higher education, connecting members from publishing, libraries, and scholarly organizations. They focus on technological innovation and knowledge sharing.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ“– ReferenceCoastRunners AIWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebCoastRunners boatOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”specification-gaminggoodharts-law+11 articleView β†’
πŸ”— WebCode of Practice on marking and labelling of AI-generated contentEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ”— WebCognitive Offloading ResearchGoogle Scholar

Research explores how humans use external resources to support cognitive tasks, examining benefits and potential limitations of this cognitive strategy.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”automationhuman-factors+11 articleView β†’
πŸ“„ PaperCohen et al. (2024)arXivβ˜…β˜…β˜…β˜†β˜†SummaryWillem Fourie2025-10-29alignmentinterpretability+31 articleView β†’
πŸ”— WebCold Takesβ€”Noneβ€”β€”effective-altruismai-safety-funding+11 articleView β†’
πŸ”— Webcollaboration between SecureBio and MITβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ“„ PaperCollective Constitutional AIAnthropic

Researchers used the Polis platform to gather constitutional principles from ~1,000 Americans. They trained a language model using these publicly sourced principles and compared it to their standard model.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”llmx-risk+34 articlesView β†’
πŸ”— WebCollective Intelligence Projectβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebColorado AI Actβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentColorado AI Act (SB 24-205)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentColorado Attorney Generalβ€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebColumbia Journalism Reviewβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCommerce Department's Framework for AI Diffusionβ€”Noneβ€”β€”compute-thresholdsgovernance+11 articleView β†’
πŸ›οΈ GovernmentCommerce Strengthens Export Controlsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentcommitmentsUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebCommunity Notes on X/Twitterβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ›οΈ GovernmentCompany earnings reportsβ€”Noneβ€”β€”economic-inequalitymarket-concentration+11 articleView β†’
πŸ“„ PaperCompetition and AI SafetyarXivβ˜…β˜…β˜…β˜†β˜†SummaryStefano Favaro, Matteo Sesia2022-09-05safetynetworks+21 articleView β†’
πŸ“„ PaperCompetition-level code generation with AlphaCodearXivβ˜…β˜…β˜…β˜†β˜†SummaryYujia Li, David Choi, Junyoung Chung, Nate Kushman, Julian Schrittwieser, RΓ©mi Leblond, Tom Eccles, James Keeling, Felix Gimeno, Agustin Dal Lago, Thomas Hubert, Peter Choy, Cyprien de Masson d'Autume, Igor Babuschkin, Xinyun Chen, Po-Sen Huang, Johannes Welbl, Sven Gowal, Alexey Cherepanov, James Molloy, Daniel J. Mankowitz, Esme Sutherland Robson, Pushmeet Kohli, Nando de Freitas, Koray Kavukcuoglu, Oriol Vinyals2022-02-08capabilitiestraining+31 articleView β†’
πŸ”— Webcomprehensive coverage from Axiosβ€”Noneβ€”β€”chatgptgpt-4+11 articleView β†’
πŸ”— Webcomprehensive documentation and transparency requirementsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ Papercomprehensive study on agent securityarXivβ˜…β˜…β˜…β˜†β˜†SummaryYifeng He, Ethan Wang, Yuyang Rong, Zifei Cheng, Hao Chen2024-06-12safetyevaluation+31 articleView β†’
πŸ”— WebComprehensive surveillance systemsRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— Webcomprehensive survey of over 250 papersβ€”Noneβ€”β€”traininghuman-feedback+12 articlesView β†’
πŸ”— WebComputational Democracy Project

The Computational Democracy Project develops Polis, an open-source platform using machine learning to understand collective group opinions. The technology enables large-scale, real-time analysis of complex group perspectives.

β€”Fullβ€”β€”open-sourceβ€”View β†’
πŸ›οΈ Governmentcompute governanceCentre for the Governance of AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecompute1 articleView β†’
✏️ Blog PostCompute governanceLessWrongβ˜…β˜…β˜…β˜†β˜†MetadataVishakha, Algon2024-12-23governancecompute1 articleView β†’
πŸ”— WebCompute Governance ReportRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecompute+32 articlesView β†’
πŸ›οΈ GovernmentCompute governance researchCentre for the Governance of AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecompute1 articleView β†’
πŸ”— WebCompute trend analysisEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”computerisk-factor+21 articleView β†’
πŸ›οΈ GovernmentCompute-Based RegulationsCentre for the Governance of AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecompute1 articleView β†’
πŸ”— WebComputer Weekly: 27 Nations and EU Statementβ€”Noneβ€”β€”compute1 articleView β†’
πŸ”— WebComputer Weekly: San Francisco Office Announcementβ€”Noneβ€”β€”compute1 articleView β†’
πŸ”— WebComputer Weekly: UK AI Safety Research Funding

The UK government established a research funding initiative to explore AI safety challenges across critical sectors. The programme aims to identify and mitigate potential risks through collaborative research grants.

β€”Fullβ€”β€”safetycomputeβ€”View β†’
πŸ”— WebComputers in Human BehaviorScienceDirectβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”computemental-health+21 articleView β†’
πŸ“„ PaperComputing Power and the Governance of AIarXiv

The paper explores how computing power can be used to enhance AI governance through visibility, resource allocation, and enforcement mechanisms. It examines the technical and policy opportunities of compute governance while also highlighting potential risks.

β˜…β˜…β˜…β˜†β˜†FullSastry, Girish, Heim, Lennart, Belfield, Haydn, Anderljung, Markus, Brundage, Miles, Hazell, Julian, O'Keefe, Cullen, Hadfield, Gillian K., Ngo, Richard, Pilz, Konstantin, Gor, George, Bluemke, Emma, Shoker, Sarah, Egan, Janet, Trager, Robert F., Avin, Shahar, Weller, Adrian, Bengio, Yoshua, Coyle, Diane2024governancecompute1 articleView β†’
πŸ“„ PaperConcrete Problems in AI SafetyarXivβ˜…β˜…β˜…β˜†β˜†SummaryDario Amodei, Chris Olah, Jacob Steinhardt, Paul Christiano, John Schulman, Dan ManΓ©2016-06-21safetyevaluation+36 articlesView β†’
πŸ“„ Paperconfidence escalation in debatesarXivβ˜…β˜…β˜…β˜†β˜†SummaryPradyumna Shyama Prasad, Minh Nhat Nguyen2025-05-25alignmentgovernance+31 articleView β†’
πŸ›οΈ GovernmentCongress.gov CRS ReportUS Congressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”compute-thresholdsgovernance+11 articleView β†’
πŸ”— WebCongressional AI briefingsCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentCongressional HearingUS Congressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”effective-altruismai-safety-funding+11 articleView β†’
πŸ›οΈ GovernmentCongressional Research Serviceβ€”Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ›οΈ GovernmentCongressional Research ServiceUS Congressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebConjecture Blogβ€”Noneβ€”β€”cognitive-emulationcoem+11 articleView β†’
πŸ”— WebConjecture Blogβ€”Noneβ€”β€”cognitive-emulationcoem+11 articleView β†’
πŸ”— WebConnected by Dataβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebConnecticut passed Senate in 2024β€”Noneβ€”β€”β€”1 articleView β†’
🎀 TalkConnor Leahy Talksβ€”Noneβ€”β€”cognitive-emulationcoem+11 articleView β†’
πŸ”— WebConservative ResearchersOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebConstitutional AI Evaluation SuiteGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”evaluation1 articleView β†’
πŸ”— WebConstitutional AI Policy BriefAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governance1 articleView β†’
πŸ”— WebConstitutional AI: AnthropicAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”agenticplanning+32 articlesView β†’
πŸ“„ PaperConstitutional AI: Harmlessness from AI FeedbackAnthropic

Anthropic introduces a novel approach to AI training called Constitutional AI, which uses self-critique and AI feedback to develop safer, more principled AI systems without extensive human labeling.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”safetytraining+311 articlesView β†’
πŸ“„ PaperConstitutional AI: Harmlessness from AI FeedbackarXivβ˜…β˜…β˜…β˜†β˜†MetadataBai, Yuntao, Kadavath, Saurav, Kundu, Sandipan, Askell, Amanda, Kernion, Jackson, Jones, Andy, Chen, Anna, Goldie, Anna, Mirhoseini, Azalia, McKinnon, Cameron, Chen, Carol, Olsson, Catherine, Olah, Christopher, Hernandez, Danny, Drain, Dawn, Ganguli, Deep, Li, Dustin, Tran-Johnson, Eli, Perez, Ethan, Kerr, Jamie, Mueller, Jared, Ladish, Jeffrey, Landau, Joshua, Ndousse, Kamal, Lukosuite, Kamile, Lovitt, Liane, Sellitto, Michael, Elhage, Nelson, Schiefer, Nicholas, Mercado, Noemi, DasSarma, Nova, Lasenby, Robert, Larson, Robin, Ringer, Sam, Johnston, Scott, Kravec, Shauna, Showk, Sheer El, Fort, Stanislav, Lanham, Tamera, Telleen-Lawton, Timothy, Conerly, Tom, Henighan, Tom, Hume, Tristan, Bowman, Samuel R., Hatfield-Dodds, Zac, Mann, Ben, Amodei, Dario, Joseph, Nicholas, McCandlish, Sam, Brown, Tom, Kaplan, Jared2022foundation-modelstransformers+320 articlesView β†’
πŸ”— WebConstitutional ClassifiersAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebConsumer Financial Protection Bureau estimatesBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ“„ Papercontamination studiesarXivβ˜…β˜…β˜…β˜†β˜†SummaryShuo Yang, Wei-Lin Chiang, Lianmin Zheng, Joseph E. Gonzalez, Ion Stoica2023-11-08capabilitiestraining+21 articleView β†’
πŸ”— WebContent Authenticity Initiative

An industry collaborative effort developing open-source tools to provide content credentials and transparency in digital media. Focuses on addressing misinformation and building trust in the age of AI-generated content.

β€”Fullβ€”β€”open-sourceβ€”View β†’
πŸ“– ReferenceContent Authenticity InitiativeWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ›οΈ GovernmentContent Credentials guidanceβ€”Noneβ€”β€”deepfakesdigital-evidence+32 articlesView β†’
πŸ”— Webcontentauthenticity.orgβ€”Noneβ€”β€”deepfakesdigital-evidence+33 articlesView β†’
πŸ”— WebContext Window Comparison 2025

ChatGPT, Claude, and Gemini are developing advanced techniques to increase context window sizes, enabling more sophisticated document analysis and reasoning across longer inputs.

β€”Fullβ€”β€”llmβ€”View β†’
πŸ”— WebControlNetGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ“„ Papercontroversial claims assessmentarXivβ˜…β˜…β˜…β˜†β˜†SummarySalman Rahman, Sheriff Issaka, Ashima Suvarna, Genglin Liu, James Shiffer, Jaeyoung Lee, Md Rizwan Parvez, Hamid Palangi, Shi Feng, Nanyun Peng, Yejin Choi, Julian Michael, Liwei Jiang, Saadia Gabriel2025-06-02capabilitiesevaluation+31 articleView β†’
πŸ”— WebCooperative AI Foundation's taxonomyβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperCooperative AI researcharXivβ˜…β˜…β˜…β˜†β˜†SummarySunil Arora, John Hastings2025-12-19governancecybersecurity+31 articleView β†’
πŸ›οΈ GovernmentCoordinated federal approachβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebCOPE (Committee on Publication Ethics)β€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebCorrigibility ResearchMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agenticplanning+38 articlesView β†’
πŸ”— WebCost-Effective Constitutional ClassifiersAnthropic Alignment

The study explores reducing computational overhead in AI safety classifiers by repurposing model computations. Methods like linear probing and fine-tuning small model sections show promising performance with minimal computational cost.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiessafety+1β€”View β†’
πŸ“„ PaperCoT Monitor+arXivβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
✏️ Blog PostCotra (2022)Alignment Forumβ˜…β˜…β˜…β˜†β˜†Metadatapaulfchristiano2021-04-07mesa-optimizationinner-alignment+11 articleView β†’
πŸ”— WebCotra (2022) - AI Takeoverβ€”Noneβ€”β€”probabilitystrategic-deception+11 articleView β†’
πŸ”— WebCould AI help bioterrorists unleash a new pandemic?β€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebCouncil of Europe Framework Conventionβ€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebCouncil of Europe: AI Treaty Portalβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCouncil on Foreign Relations analysisβ€”Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ”— WebCoursera AI governanceβ€”Noneβ€”β€”governance1 articleView β†’
πŸ”— WebCox & Palmer AIDA Analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCraig Newmark Philanthropiesβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentCREATE AI ActUS Congressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ“„ PaperCross-cultural study, 2024arXivβ˜…β˜…β˜…β˜†β˜†SummaryHaijiang Liu, Jinguang Gu, Xun Wu, Daniel Hershcovich, Qiaoling Xiao2025-11-21alignmentinterpretability+31 articleView β†’
πŸ”— WebCrossref Event Dataβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebCrowdStrikeβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ›οΈ GovernmentCRSUS Congressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCrunchbase - 6 Charts That Show The Big AI Funding Trends Of 2025

Crunchbase data reveals AI captured nearly 50% of global startup funding in 2025, with $202.3 billion invested. Foundation model companies like OpenAI and Anthropic attracted the largest investments.

β€”Fullβ€”β€”β€”2 articlesView β†’
πŸ”— WebCSET analysisCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebCSET analysisCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebCSET GeorgetownCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”scalingcapability-evaluation+12 articlesView β†’
πŸ”— WebCSET Georgetown - The AI Competition with ChinaCSET Georgetown

Examines the AI technological and strategic competition between the United States and China, focusing on diplomatic strategies and potential risks in AI development.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebCSET Georgetown researchCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCSET researchβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCSET researchCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCSET ResearchCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebCSET: AI Market DynamicsCSET Georgetown

I apologize, but the provided content appears to be a fragmentary collection of references or headlines rather than a substantive document that can be comprehensively analyzed. Without a complete, coherent source text, I cannot generate a meaningful summary or review. To properly complete the task, I would need: 1. A full research document or article 2. Clear contextual content explaining the research's scope, methodology, findings 3. Sufficient detail to extract meaningful insights If you have the complete source document, please share it and I'll be happy to provide a thorough analysis following the specified JSON format. Would you like to: - Provide the full source document - Clarify the source material - Select a different document for analysis

β˜…β˜…β˜…β˜…β˜†Summaryβ€”β€”prioritizationresource-allocation+319 articlesView β†’
πŸ”— WebCSIS analysisCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCSIS analysisCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ”— WebCSIS analysisCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCSIS analystsCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebCSIS Big Data Chinaβ€”Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ”— WebCSIS Big Data China Projectβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebCSIS Critical QuestionsCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theoryinternational-coordination+33 articlesView β†’
πŸ”— WebCSIS economic security analysisCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economiccybersecurity+31 articleView β†’
πŸ”— WebCSIS Mismatch reportCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCSIS reportedCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCSIS: G7 Hiroshima AI ProcessCSIS

The report examines the G7's emerging approach to AI governance, highlighting potential enhancements for international cooperation on AI development and regulation.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governance1 articleView β†’
πŸ”— WebCSIS: UN Global Dialogue on AI GovernanceCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governance2 articlesView β†’
πŸ”— WebCSIS: Understanding Biden Administration Export ControlsCSIS

I apologize, but the provided source does not appear to be a comprehensive document about AI safety. While it seems to reference export controls related to the Biden administration, the text appears to be incomplete or a header/introduction rather than a full research document. Without the full content, I cannot responsibly generate a comprehensive summary. To properly complete this task, I would need: 1. The full text of the document 2. Clear sections discussing the research findings 3. Methodological details 4. Conclusions and implications If you have the complete document, I'm happy to analyze it using the requested JSON format. Otherwise, I cannot fabricate details about a partial or missing source. Would you like to provide the complete document text?

β˜…β˜…β˜…β˜…β˜†Summaryβ€”β€”safety1 articleView β†’
πŸ”— WebCSIS: US Vision for AI SafetyCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety2 articlesView β†’
πŸ”— WebCSMonitorβ€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebCuban Missile Crisisβ€”Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— WebCurrent compute trendsEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”computetimeline+21 articleView β†’
πŸ”— WebCursor Agent Modeβ€”Noneβ€”β€”agenticplanning+11 articleView β†’
πŸ”— WebCVPR 2024β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebCyberpsychology, Behavior, and Social Networkingβ€”Noneβ€”β€”cybersecuritymental-health+21 articleView β†’
πŸ”— WebCyberSecEvalGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”cybersecuritysoftware-engineering+21 articleView β†’
πŸ”— WebCybersecurity implications of AICSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecuritytimeline+21 articleView β†’
πŸ”— WebCybersecurity Ventures projectsβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebCyberSeekβ€”Noneβ€”β€”cybersecurity1 articleView β†’
πŸ”— Webcylab.cmu.edu

CyLab coordinates security and privacy research across Carnegie Mellon University departments, promoting collaborative research and education. The institute aims to drive significant impact in security research, policy, and practice.

β€”Fullβ€”β€”governancecybersecurityβ€”View β†’
πŸ”— WebCymulateβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ“„ PaperDafoe (2018)arXivβ˜…β˜…β˜…β˜†β˜†SummaryC. Gauvin-Ndiaye, T. E. Baker, P. Karan, Γ‰. MassΓ©, M. Balli, N. Brahiti, M. A. Eskandari, P. Fournier, A. -M. S. Tremblay, R. Nourafkan2018-09-20resource-allocationresearch-priorities+11 articleView β†’
πŸ”— WebDALL-E 3OpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebDan HendrycksTIMEβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”evaluationsdangerous-capabilities+32 articlesView β†’
πŸ“„ PaperDangerous Capability EvaluationsarXivβ˜…β˜…β˜…β˜†β˜†SummaryMary Phuong, Matthew Aitchison, Elliot Catt, Sarah Cogan, Alexandre Kaskasoli, Victoria Krakovna, David Lindner, Matthew Rahtz, Yannis Assael, Sarah Hodkinson, Heidi Howard, Tom Lieberum, Ramana Kumar, Maria Abi Raad, Albert Webson, Lewis Ho, Sharon Lin, Sebastian Farquhar, Marcus Hutter, Gregoire Deletang, Anian Ruoss, Seliem El-Sayed, Sasha Brown, Anca Dragan, Rohin Shah, Allan Dafoe, Toby Shevlane2024-03-20capabilitiessafety+31 articleView β†’
πŸ”— WebDaniel Kokotajlo reveals ~50% AGI safety staff departedFortuneβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safetyagi1 articleView β†’
πŸ”— WebDario AmodeiAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDario AmodeiAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDario AmodeiAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorcompetition+11 articleView β†’
πŸ”— WebDario Amodeiβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDario Amodei (Anthropic)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDario Amodei notedβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDario Amodei's analysisAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebDarknet Diaries: Voice Phishing Episodesβ€”Noneβ€”β€”cybersecurityinformation-warfare+32 articlesView β†’
πŸ”— WebDaron Acemoglu (MIT)β€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebDARPAβ€”Noneβ€”β€”escalationconflict+35 articlesView β†’
πŸ”— WebDARPA Cyber Grand Challengeβ€”Noneβ€”β€”cybersecuritytimeline+11 articleView β†’
πŸ”— WebDARPA MediFor Program

DARPA's MediFor program addresses the challenge of image manipulation by developing advanced forensic technologies to assess visual media integrity. The project seeks to create an automated platform that can detect and analyze digital image and video alterations.

β€”Fullβ€”β€”economicepistemic+35 articlesView β†’
πŸ”— WebDARPA SemaFor

SemaFor focuses on creating advanced detection technologies that go beyond statistical methods to identify semantic inconsistencies in deepfakes and AI-generated media. The program aims to provide defenders with tools to detect manipulated content across multiple modalities.

β€”Fullβ€”β€”deepfakescontent-verification+12 articlesView β†’
πŸ”— WebDARPA's researchβ€”Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— WebDarwin Godel Machineβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebData & Societyβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebData & Society: Alternative Influenceβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebData Security Lawβ€”Noneβ€”β€”cybersecurityregulation+21 articleView β†’
πŸ”— WebData-Centric Authoritarianism

The report examines how China is developing advanced technologies like AI surveillance, neurotechnologies, quantum computing, and digital currencies that enable unprecedented data collection and social control. These technologies pose significant risks to privacy and democratic freedoms.

β€”Fullβ€”β€”x-riskgovernance+12 articlesView β†’
πŸ”— WebDataCamp ML Salariesβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDavidson & Houlden 2025β€”Noneβ€”β€”intelligence-explosionrecursive-self-improvement+11 articleView β†’
πŸ“„ PaperDe Fauw et al. (2018)Natureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ”— WebDebateβ€”Noneβ€”β€”traininghuman-feedback+11 articleView β†’
πŸ“„ PaperDebatearXivβ˜…β˜…β˜…β˜†β˜†SummaryZachary Kenton, Noah Y. Siegel, JΓ‘nos KramΓ‘r, Jonah Brown-Cohen, Samuel Albanie, Jannis Bulian, Rishabh Agarwal, David Lindner, Yunhao Tang, Noah D. Goodman, Rohin Shah2024-07-05capabilitiesevaluation+12 articlesView β†’
πŸ“„ PaperDebate as Scalable OversightarXivβ˜…β˜…β˜…β˜†β˜†SummaryGeoffrey Irving, Paul Christiano, Dario Amodei2018-05-02alignmentsafety+38 articlesView β†’
πŸ”— WebDebate May Help AI Models Converge on Truthβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperDebate, Deliberate, Decide (D3): Cost-Aware Adversarial FrameworkarXivβ˜…β˜…β˜…β˜†β˜†SummaryChaithanya Bandi, Abir Harrasse2024-10-07interpretabilityevaluation+31 articleView β†’
πŸ”— WebDecember 2024 first fully unmanned operation near LyptsiAtlantic Councilβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebDecember 2024 UN General Assembly resolutionβ€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebDeceptively Aligned Mesa-Optimizersβ€”Noneβ€”β€”alignmentdeception+11 articleView β†’
πŸ”— WebDecidim softwareβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebDeciphering Russia's "Sovereign Internet Law"β€”Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ”— WebDecision Lab: Autonomy in AI-Driven Futureβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebDeclining physician trustβ€”Noneβ€”β€”β€”β€”View β†’
✏️ Blog PostDecoupling Deliberation and DeploymentAlignment Forumβ˜…β˜…β˜…β˜†β˜†Metadatapaulfchristiano2018-05-25β€”1 articleView β†’
πŸ“„ PaperDeep LearningNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebDeep Learning textbook (2016)β€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ›οΈ GovernmentDeep Synthesis Provisionsβ€”Noneβ€”β€”regulationchina+11 articleView β†’
πŸ”— WebDeepFaceLiveGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ“„ PaperDeepfake attempts increased 3,000% in 2023β€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperDeepfake detection accuracy decliningarXiv

A survey exploring the creation and detection of deepfakes, examining technological advancements, current trends, and potential threats in generative AI technologies.

β˜…β˜…β˜…β˜†β˜†FullMirsky, Yisroel, Lee, Wenke2020deepfakescontent-verification+32 articlesView β†’
πŸ“„ PaperDeepfake-Eval-2024 benchmarkarXivβ˜…β˜…β˜…β˜†β˜†SummaryNuria Alina Chandra, Ryan Murtfeldt, Lin Qiu, Arnab Karmakar, Hannah Lee, Emmanuel Tanumihardja, Kevin Farhat, Ben Caffee, Sejin Paik, Changyeon Lee, Jongwook Choi, Aerin Kim, Oren Etzioni2025-03-04capabilitiesevaluation+31 articleView β†’
πŸ“„ PaperDeepfake-Eval-2024 BenchmarkarXivβ˜…β˜…β˜…β˜†β˜†SummaryNuria Alina Chandra, Ryan Murtfeldt, Lin Qiu, Arnab Karmakar, Hannah Lee, Emmanuel Tanumihardja, Kevin Farhat, Ben Caffee, Sejin Paik, Changyeon Lee, Jongwook Choi, Aerin Kim, Oren Etzioni2025-03-04capabilitiesevaluation+32 articlesView β†’
πŸ”— Webdeepfakedetectionchallenge.aiβ€”Noneβ€”β€”historical-evidencearchives+11 articleView β†’
πŸ”— WebDeepfakes research by the University of Washingtonβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebDeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”foundation-modelstransformers+32 articlesView β†’
πŸ”— WebDeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”agenticplanning+11 articleView β†’
πŸ”— WebDeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interventionseffectiveness+33 articlesView β†’
πŸ”— WebDeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”networksrisk-interactions+32 articlesView β†’
πŸ”— WebDeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”resource-allocationresearch-priorities+11 articleView β†’
πŸ”— WebDeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cost-effectivenessresearch-priorities+11 articleView β†’
✏️ Blog PostDeepMindβ€”Noneβ€”β€”interpretabilityscalable-oversight+11 articleView β†’
πŸ”— WebDeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interpretabilityscalable-oversight+11 articleView β†’
πŸ”— WebDeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDeepMindβ€”Noneβ€”β€”β€”1 articleView β†’
✏️ Blog PostDeepMind alignment agendaLessWrongβ˜…β˜…β˜…β˜†β˜†MetadataVika2018-11-20alignmentdebate+22 articlesView β†’
πŸ“„ PaperDeepMind Cicero researchScienceβ˜…β˜…β˜…β˜…β˜…SummaryA. Bakhtin, Noam Brown, Emily Dinan, Gabriele Farina, Colin Flaherty, Daniel Fried, A. Goff, Jonathan Gray, Hengyuan Hu, Athul Paul Jacob, Mo-jtaba Komeili, Karthik Konath, Minae Kwon, Adam Lerer, Mike Lewis, Alexander H. Miller, S. Mitts, Adithya Renduchintala, Stephen Roller, Dirk Rowe, Weiyan Shi, Joe Spisak, Alexander Wei, David J. Wu, Hugh Zhang, Markus Zijlstra2022-11-22alignmentcapabilities+21 articleView β†’
πŸ”— WebDeepMind Frontier Safety FrameworkGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetymonitoring+35 articlesView β†’
πŸ”— WebDeepMind PrinciplesGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDeepMind researchβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDeepMind SafetyGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyprobability+21 articleView β†’
πŸ”— WebDeepMind Safetyβ€”Noneβ€”β€”safetycapability-generalization+21 articleView β†’
πŸ”— WebDeepMind salary dataβ€”Noneβ€”β€”research-agendasalignment+11 articleView β†’
πŸ”— WebDeepMind: AI SafetyGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyβ€”View β†’
πŸ”— WebDeepMind: An Approach to Technical AGI Safety and Securityβ€”Noneβ€”β€”safetycybersecurity+1β€”View β†’
πŸ”— WebDeepMind: Deepening AI Safety Research with UK AISIGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebDeepMind's game theory researchGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”causal-modelcorrigibility+11 articleView β†’
πŸ”— WebDeepMind's specification gaming researchGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”probabilitygeneralization+11 articleView β†’
πŸ”— WebDeepSeekMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDeepSeek R1 releaseGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”open-sourcedecision-theory+32 articlesView β†’
πŸ”— WebDeepSeek warningβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDeepSeek, Huawei, Export Controls, and the Future of the U.S.-China AI RaceCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebDeepstrikeβ€”Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ”— WebDeepstrikeβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— Webdefines slow takeoffβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webdefinition of "algorithmic discrimination"β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webdelayed release of Llama BehemothCNBCβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”open-source1 articleView β†’
πŸ“„ Paperdelegation chainsarXivβ˜…β˜…β˜…β˜†β˜†SummaryTobin South, Samuele Marro, Thomas Hardjono, Robert Mahari, Cedric Deslandes Whitney, Dazza Greenwood, Alan Chan, Alex Pentland2025-01-16capabilitiescybersecurity1 articleView β†’
πŸ›οΈ GovernmentDelete Actβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ“„ PaperDelfino: "Deepfakes on Trial"SSRNβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ”— WebDeliberative Pollingβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebDeloitte's 2024 analysisβ€”Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebDemandSage

Nearly 90% of companies worldwide are integrating AI technologies, with significant adoption in customer service, business operations, and strategic planning. The AI market is expected to reach $294.16 billion by 2025.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebDemandSage

Comprehensive analysis of global AI market growth, market share, adoption rates, and economic impacts across industries and regions. Highlights rapid expansion and transformative potential of artificial intelligence technologies.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebDemis Hassabis - Britannicaβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDemis Hassabis - Google DeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“– ReferenceDemis Hassabis - WikipediaWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebDemocracy and TechnologyCambridge University Pressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebDemocracy Fundβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— Webdemonstrated creating a fully AI-generated exploit for CVE-2025-32433β€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
✏️ Blog PostDenison et al. (2024)Alignment Forumβ˜…β˜…β˜…β˜†β˜†MetadataKei Nishimura-Gasparian, Isaac Dunn, Henry Sleight, Miles Turpin, evhub, Carson Denison, Ethan Perez2024-05-28traininghuman-feedback+12 articlesView β†’
πŸ”— WebDepartment of Commerce's proposed ruleβ€”Noneβ€”β€”compute-thresholdsgovernance+12 articlesView β†’
✏️ Blog Postdeprioritizing SAE researchβ€”Noneβ€”β€”sparse-autoencodersfeatures+11 articleView β†’
πŸ“„ PaperDesigning Incident Reporting SystemsarXivβ˜…β˜…β˜…β˜†β˜†SummaryKevin Wei, Lennart Heim2025-11-08governancesafety1 articleView β†’
πŸ”— WebDeskilling LiteratureGoogle Scholar

Deskilling literature explores how technology transforms work by reducing skill complexity and changing labor requirements across different industries.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”economicautomation+21 articleView β†’
πŸ“„ PaperDetection accuracy drops with newer generatorsarXivβ˜…β˜…β˜…β˜†β˜†SummaryNam Hyeon-Woo, Kim Yu-Ji, Byeongho Heo, Dongyoon Han, Seong Joon Oh, Tae-Hyun Oh2022-10-16capabilitiesllm+31 articleView β†’
πŸ”— WebDetection research laggingβ€”Noneβ€”β€”timelinecapability+11 articleView β†’
πŸ“„ PaperDetection tools unreliableNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— Webdeveloping national implementation plansβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“‹ ReportDeveloping Near-Miss Reporting System

A multi-pronged research project investigated near-miss reporting systems for roadside responders, examining existing platforms, stakeholder perspectives, and barriers to reporting to develop comprehensive recommendations.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebDevinβ€”Noneβ€”β€”agenticplanning+11 articleView β†’
πŸ“„ PaperDFDC Challenge resultsarXivβ˜…β˜…β˜…β˜†β˜†SummaryBrian Dolhansky, Joanna Bitton, Ben Pflaum, Jikuo Lu, Russ Howes, Menglin Wang, Cristian Canton Ferrer2020-06-12trainingevaluation+31 articleView β†’
πŸ”— WebDGAP AI/Democracy Initiativeβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebDifferential privacyMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— Webdigital signaturesβ€”Noneβ€”β€”social-engineeringvoice-cloning+11 articleView β†’
πŸ“„ PaperDirect Preference OptimizationarXivβ˜…β˜…β˜…β˜†β˜†SummaryRafael Rafailov, Archit Sharma, Eric Mitchell, Stefano Ermon, Christopher D. Manning, Chelsea Finn2023-05-29governancetraining+31 articleView β†’
πŸ”— Webdirected by the Commerce Departmentβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webdisbanded another safety teamCNBCβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safetyresearch-agendas+23 articlesView β†’
πŸ“„ PaperDistributed AI Safety (Amodei et al.)arXivβ˜…β˜…β˜…β˜†β˜†SummaryEmmanuel Klu, Sameer Sethi2023-09-07capabilitiessafety+31 articleView β†’
πŸ”— WebDLA Piper: California's SB-1047: Understanding the Safe and Secure Innovation for Frontier AI Actβ€”Noneβ€”β€”safetyregulation+21 articleView β†’
πŸ”— WebDNA Script SYNTAX Systemβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebDNA synthesis screeningβ€”Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ“„ PaperDo AI Companies Make Good on Voluntary Commitments to the White House?arXiv

Research analyzed 16 AI companies' compliance with White House voluntary AI commitments in 2023, finding wide disparities in performance with an average score of 53% and significant weaknesses in model weight security and third-party reporting.

β˜…β˜…β˜…β˜†β˜†FullWang, Jennifer, Huang, Kayla, Klyman, Kevin, Bommasani, Rishi2025capabilitiescybersecurity1 articleView β†’
πŸ“„ Paperdocumented capabilitiesarXivβ˜…β˜…β˜…β˜†β˜†SummaryXingyu Zhu, Shuo Wang, Jinda Lu, Yanbin Hao, Haifeng Liu, Xiangnan He2024-03-23interpretabilitycapabilities+31 articleView β†’
πŸ”— WebDocumented incidentsβ€”Noneβ€”β€”automationhuman-factors+11 articleView β†’
πŸ›οΈ GovernmentDoD reportsβ€”Noneβ€”β€”automationhuman-factors+11 articleView β†’
πŸ›οΈ GovernmentDOE data center report

A Department of Energy report highlights significant growth in data center energy usage, with electricity consumption expected to increase dramatically by 2028 due to AI and technological advances.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentDOJ AI Probeβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ›οΈ GovernmentDraft measuresβ€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebDreber et al. (2015)PNASβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebDwarkesh Podcastβ€”Noneβ€”β€”constitutional-airesponsible-scaling+11 articleView β†’
πŸ”— WebDwarkesh Podcast 2024β€”Noneβ€”β€”constitutional-airesponsible-scaling+11 articleView β†’
πŸ”— WebEA ForumEA Forumβ˜…β˜…β˜…β˜†β˜†Metadata80000_Hours, poppinfresh2024-10-29x-riskirreversibility+11 articleView β†’
✏️ Blog PostEA Forum analysisEA Forumβ˜…β˜…β˜…β˜†β˜†MetadataChristopher Clay2025-08-30field-buildingtraining-programs+11 articleView β†’
✏️ Blog PostEA Forum Career PostsEA Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”talentfield-building+34 articlesView β†’
πŸ”— WebEA Forum surveysEA Forumβ˜…β˜…β˜…β˜†β˜†Metadatabmg2022-06-19β€”1 articleView β†’
✏️ Blog PostEA Forum: I read every major AI lab's safety plan so you don't have toEA Forumβ˜…β˜…β˜…β˜†β˜†Metadatasarahhw2024-12-16safety2 articlesView β†’
πŸ”— WebEA Forum: Incident Reporting for AI SafetyEA Forum

The document argues for developing a comprehensive incident reporting system for AI, emphasizing the importance of sharing information about AI system failures, near-misses, and potential risks to improve overall AI safety and accountability.

β˜…β˜…β˜…β˜†β˜†FullZach Stein-Perlman, SeLo, stepanlos, MvKπŸ”Έ2023-07-19safety1 articleView β†’
✏️ Blog PostEA Forum: Is Pausing AI Possible?EA Forumβ˜…β˜…β˜…β˜†β˜†MetadataRichard Annilo2024-10-09β€”1 articleView β†’
✏️ Blog PostEA Forum: What are Responsible Scaling Policies (RSPs)?EA Forumβ˜…β˜…β˜…β˜†β˜†MetadataVishakha Agrawal, Algon2025-04-05capabilities1 articleView β†’
πŸ”— WebEarly Best Practices for Frontier AI Safety Evaluationsβ€”Noneβ€”β€”safetyevaluation1 articleView β†’
πŸ”— WebEarly compliance effortsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentEarly stagesUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factordiffusion+12 articlesView β†’
πŸ”— Webec.europa.euβ€”Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— WebEchoLeak exploit (CVE-2025-32711)β€”Noneβ€”β€”computer-usefunction-calling+11 articleView β†’
πŸ”— WebEconomic Journalβ€”Noneβ€”β€”economicinstitutional-trust+21 articleView β†’
πŸ”— WebEdelman Trust BarometerEdelmanβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebEdelman Trust BarometerEdelmanβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”information-overloadmedia-literacy+11 articleView β†’
πŸ”— WebEdelman Trust BarometerEdelmanβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”information-overloadmedia-literacy+11 articleView β†’
πŸ“‹ ReportEdelman Trust Barometer 2024 - AI InsightsEdelmanβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebEdelman's 2024 Trust BarometerEdelmanβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebEdison and Black

AI is expected to generate millions of new jobs while transforming existing roles. Strategic upskilling and workforce development are essential to navigating this technological shift.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebEducational Psychologyβ€”Noneβ€”β€”human-agencyautomation+11 articleView β†’
πŸ”— WebEducational Psychology StudiesScienceDirectβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-agencyautomation+11 articleView β†’
πŸ”— WebEdWeek AI Surveyβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ”— WebEFF Surveillance Explainersβ€”Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebElectronic Frontier Foundationβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebEleutherAI Evaluationβ€”Noneβ€”β€”evaluationframework+34 articlesView β†’
πŸ”— WebEleven Labs' roadmapβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebElevenLabsβ€”Noneβ€”β€”capabilitythreshold+34 articlesView β†’
πŸ”— WebElicitβ€”Noneβ€”β€”intelligence-explosionrecursive-self-improvement+11 articleView β†’
πŸ”— Webeliciting latent knowledgeβ€”Noneβ€”β€”eliciting-latent-knowledgeelk+32 articlesView β†’
πŸ“– ReferenceEliezer YudkowskyWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebElizabeth KellyTIMEβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperElliott Thornley's 2024 paper "The Shutdown Problem"Springerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”corrigibilityshutdown-problem+11 articleView β†’
πŸ”— Webembedded agencyMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
✏️ Blog PostEmbedded AgencyAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agent-foundationsdecision-theory+13 articlesView β†’
πŸ“„ PaperEmergent AbilitiesarXivβ˜…β˜…β˜…β˜†β˜†SummaryJason Wei, Yi Tay, Rishi Bommasani, Colin Raffel, Barret Zoph, Sebastian Borgeaud, Dani Yogatama, Maarten Bosma, Denny Zhou, Donald Metzler, Ed H. Chi, Tatsunori Hashimoto, Oriol Vinyals, Percy Liang, Jeff Dean, William Fedus2022-06-15capabilitiesllm+34 articlesView β†’
πŸ“„ PaperEmergent capability detectionarXivβ˜…β˜…β˜…β˜†β˜†SummarySamir Yitzhak Gadre, Gabriel Ilharco, Alex Fang, Jonathan Hayase, Georgios Smyrnis, Thao Nguyen, Ryan Marten, Mitchell Wortsman, Dhruba Ghosh, Jieyu Zhang, Eyal Orgad, Rahim Entezari, Giannis Daras, Sarah Pratt, Vivek Ramanujan, Yonatan Bitton, Kalyani Marathe, Stephen Mussmann, Richard Vencu, Mehdi Cherti, Ranjay Krishna, Pang Wei Koh, Olga Saukh, Alexander Ratner, Shuran Song, Hannaneh Hajishirzi, Ali Farhadi, Romain Beaumont, Sewoong Oh, Alex Dimakis, Jenia Jitsev, Yair Carmon, Vaishaal Shankar, Ludwig Schmidt2023-04-27capabilitiestraining+31 articleView β†’
πŸ”— WebEmergingβ€”Noneβ€”β€”risk-factorgame-theory+11 articleView β†’
πŸ”— WebEmerging Technology Observatory - State of Global AI Safety Research

An analysis of global AI safety research trends from 2017-2022 reveals significant growth and American leadership in the field. The research examines publication volumes, citations, and key research clusters.

β€”Fullβ€”β€”safetyβ€”View β†’
πŸ”— WebEmily Bender's work

Emily Bender is a University of Washington linguistics professor who researches computational linguistics, grammar engineering, and the ethical implications of language technologies. Her work critically examines the societal impacts of natural language processing and AI systems.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebEndsley (2017)β€”Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ“„ PaperEngineered prompts in emailsarXivβ˜…β˜…β˜…β˜†β˜†SummaryShrestha Datta, Shahriar Kabir Nahin, Anshuman Chhabra, Prasant Mohapatra2025-10-27governancecapabilities+31 articleView β†’
πŸ“„ PaperEnsemble methods researcharXivβ˜…β˜…β˜…β˜†β˜†SummaryDan Hendrycks, Norman Mu, Ekin D. Cubuk, Barret Zoph, Justin Gilmer, Balaji Lakshminarayanan2019-12-05capabilitiestraining+31 articleView β†’
πŸ”— WebEPA rule commentsβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebEpoch AIEpoch AI

Epoch AI provides comprehensive data and insights on AI model scaling, tracking computational performance, training compute, and model developments across various domains.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiestraining+314 articlesView β†’
πŸ”— WebEpoch AIEpoch AI

Epoch AI is a research organization collecting and analyzing data on AI model training compute, computational performance, and technological trends in artificial intelligence.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiestraining+1β€”View β†’
πŸ”— WebEpoch AIEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+12 articlesView β†’
πŸ”— WebEpoch AIEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+12 articlesView β†’
πŸ”— WebEpoch AIEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+12 articlesView β†’
πŸ”— WebEpoch AIEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorcompetition+11 articleView β†’
πŸ”— WebEpoch AIEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-forecastingcompute-trends+11 articleView β†’
πŸ”— WebEpoch AIEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”3 articlesView β†’
πŸ”— WebEpoch AI (2024)Epoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cost-effectivenessresearch-priorities+11 articleView β†’
πŸ”— WebEpoch AI 2024Epoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ”— WebEpoch AI algorithmic progressEpoch AI

A comprehensive analysis of language model algorithmic progress reveals rapid efficiency improvements, with compute requirements halving approximately every 8 months. However, compute scaling contributes 60-95% of performance improvements.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiescompute+11 articleView β†’
πŸ”— WebEpoch AI computing capacityEpoch AI

Epoch AI analyzed computing capacity across leading tech companies, estimating their AI chip holdings in H100 equivalents. Google, Microsoft, Meta, and Amazon collectively own substantial AI computing power, primarily through NVIDIA and Google's TPU chips.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”compute1 articleView β†’
πŸ”— WebEpoch AI consumer GPU analysisEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”compute1 articleView β†’
πŸ”— WebEpoch AI estimatesEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”regulationgpai+11 articleView β†’
πŸ”— WebEpoch AI GPU production trackingEpoch AI

Epoch AI tracked NVIDIA GPU computing power growth, finding a 2.3x annual increase since 2019. The Hopper generation currently dominates with 77% of total AI hardware computing power.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”compute1 articleView β†’
πŸ”— WebEpoch AI inference allocationEpoch AI

A theoretical analysis suggests that the most efficient compute spending for AI models involves approximately equal investment in training and inference, with techniques like pruning and sampling allowing compute trade-offs.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”trainingcompute1 articleView β†’
πŸ”— WebEpoch AI model databaseEpoch AI

Epoch AI analyzed the landscape of large-scale AI models, identifying over 30 models trained with more than 10^25 floating-point operations (FLOP). The analysis covers models from leading AI developers across language, reasoning, and multimodal domains.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebEpoch AI OpenAI compute spendEpoch AI

Epoch AI analyzed OpenAI's 2024 compute spending, estimating $5 billion in R&D compute and $2 billion in inference compute. Most compute was likely used for experimental and unreleased model training.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”trainingcompute1 articleView β†’
πŸ”— WebEpoch AI power analysisEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebEpoch AI projectionsEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+12 articlesView β†’
πŸ”— WebEpoch AI researchEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”regulationgpai+11 articleView β†’
πŸ”— WebEpoch AI research from October 2025Epoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebEpoch AI trackingEpoch AI

Epoch AI presents a comprehensive dataset tracking the development of large-scale AI models, showing exponential growth in training compute and model complexity across various domains.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”trainingcompute1 articleView β†’
πŸ”— WebEpoch AI training costsEpoch AI

A comprehensive study examining the dollar cost of training machine learning systems shows training costs have been increasing by around 0.5 orders of magnitude annually, with significant uncertainties and variations between different types of systems.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”training1 articleView β†’
πŸ”— WebEpoch AI: Literature Review of TAI TimelinesEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebEpoch AI's analysisEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alphafolddrug-discovery+12 articlesView β†’
πŸ”— WebEpoch's compute databaseEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”computeai-forecasting+21 articleView β†’
πŸ”— Webepochai.org/blogEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-forecastingcompute-trends+11 articleView β†’
πŸ”— Webepochai.org/researchEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-forecastingcompute-trends+11 articleView β†’
πŸ”— WebEpstein & Robertson (2015)PNASβ˜…β˜…β˜…β˜…β˜…MetadataEpstein, Robert, Robertson, Ronald E.2015ai-ethicspersuasion+11 articleView β†’
πŸ“„ PaperErdil & Besiroglu (2023)arXivβ˜…β˜…β˜…β˜†β˜†SummarySarah Gao, Andrew Kean Gao2023-07-19trainingllm1 articleView β†’
πŸ“– ReferenceEric SchmidtWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ Governmentestablished in February 2024β€”Noneβ€”β€”governancegovernment-oversight+12 articlesView β†’
πŸ“„ PaperEsvelt - Delay, Detect, Defend (2022)β€”Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ”— WebEthereum: Futarchy experiments

Futarchy is a governance model where participants bet on potential policy outcomes using prediction markets, with the goal of selecting policies that maximize a predefined success metric. It aims to leverage market dynamics to make more rational and effective organizational decisions.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ”— WebETSI TC SAIβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebETSI TS 104 223β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebEU AI Act

The EU AI Act introduces the world's first comprehensive AI regulation, classifying AI applications into risk categories and establishing legal frameworks for AI development and deployment.

β€”Fullβ€”β€”governancesoftware-engineering+318 articlesView β†’
πŸ”— WebEU AI Actβ€”Noneβ€”β€”risk-interactionscompounding-effects+11 articleView β†’
πŸ”— WebEU AI Actβ€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebEU AI Actβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebEU AI Actβ€”Noneβ€”β€”β€”3 articlesView β†’
πŸ”— WebEU AI Actβ€”Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ”— WebEU AI Actβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebEU AI Act Article 5β€”Noneβ€”β€”regulationgpai+11 articleView β†’
πŸ”— WebEU AI Act Article 50β€”Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ”— WebEU AI Act enters into forceβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebEU AI Act Implementation Timeline

The EU AI Act implementation follows a gradual rollout with key dates from 2024 to 2031, establishing progressive regulatory milestones for AI systems and governance.

β€”Fullβ€”β€”governance4 articlesView β†’
πŸ”— WebEU AI Act Implementation Timelineβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebEU AI Act provisionsβ€”Noneβ€”β€”probabilitydecomposition+33 articlesView β†’
πŸ”— WebEU AI Act StandardisationEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebEU AI Liability DirectiveEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ”— WebEU AI Liability Directiveβ€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebEU AI OfficeEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+311 articlesView β†’
πŸ”— WebEU AI OfficeEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theorygovernance+11 articleView β†’
πŸ”— WebEU AI OfficeEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ”— WebEU Code of Conductβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebEU Commission DG COMPβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ”— WebEU Conference on Future of Europeβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebEU Digital Services ActEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-ai-interactionai-control+33 articlesView β†’
πŸ”— WebEU Financial Instruments Directiveβ€”Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebEU High-Level Expert Group on AIEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebEU placed Tevian and NtechLab under sanctionsβ€”Noneβ€”β€”authoritarianismstability+11 articleView β†’
πŸ”— WebEU tech regulationβ€”Noneβ€”β€”governanceeconomic-inequality+21 articleView β†’
πŸ”— WebEU's Digital Services Actβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— Webeur-lex.europa.euβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebEuronews: International AI Treaty Signing

The AI Treaty provides a comprehensive legal framework for AI system regulation across public and private sectors. It allows non-EU countries to sign and aims to promote responsible AI innovation while addressing potential risks.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ”— WebEuropean Commission impact assessmentEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”regulationgpai+11 articleView β†’
πŸ”— WebEuropean Commission: EU AI ActEuropean Union

The EU AI Act is a pioneering legal framework classifying AI systems by risk levels and setting strict rules for high-risk and potentially harmful AI applications to protect fundamental rights and ensure safety.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”safetygame-theory+36 articlesView β†’
πŸ”— WebEuropean Council on Foreign Relations researchβ€”Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ”— WebEuropean Media and Information Fundβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebEuropean Parliament studyβ€”Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ”— WebEuropean Parliament: EU AI Act Overview

The EU AI Act establishes a comprehensive regulatory framework for artificial intelligence, classifying AI systems by risk levels and imposing transparency and safety requirements.

β€”Fullβ€”β€”governancesafety+32 articlesView β†’
πŸ”— WebEuropol reportβ€”Noneβ€”β€”truthepistemology+11 articleView β†’
πŸ”— WebEurostat dataβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebEvalPlusβ€”Noneβ€”β€”evaluation1 articleView β†’
πŸ“„ PaperEvaluating Frontier Models for Dangerous CapabilitiesarXivβ˜…β˜…β˜…β˜†β˜†SummaryMary Phuong, Matthew Aitchison, Elliot Catt, Sarah Cogan, Alexandre Kaskasoli, Victoria Krakovna, David Lindner, Matthew Rahtz, Yannis Assael, Sarah Hodkinson, Heidi Howard, Tom Lieberum, Ramana Kumar, Maria Abi Raad, Albert Webson, Lewis Ho, Sharon Lin, Sebastian Farquhar, Marcus Hutter, Gregoire Deletang, Anian Ruoss, Seliem El-Sayed, Sasha Brown, Anca Dragan, Rohin Shah, Allan Dafoe, Toby Shevlane2024-03-20capabilitiessafety+31 articleView β†’
πŸ“„ PaperEvaluating Large Language Models Trained on CodearXivβ˜…β˜…β˜…β˜†β˜†SummaryMark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde de Oliveira Pinto, Jared Kaplan, Harri Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman, Alex Ray, Raul Puri, Gretchen Krueger, Michael Petrov, Heidy Khlaaf, Girish Sastry, Pamela Mishkin, Brooke Chan, Scott Gray, Nick Ryder, Mikhail Pavlov, Alethea Power, Lukasz Kaiser, Mohammad Bavarian, Clemens Winter, Philippe Tillet, Felipe Petroski Such, Dave Cummings, Matthias Plappert, Fotios Chantzis, Elizabeth Barnes, Ariel Herbert-Voss, William Hebgen Guss, Alex Nichol, Alex Paino, Nikolas Tezak, Jie Tang, Igor Babuschkin, Suchir Balaji, Shantanu Jain, William Saunders, Christopher Hesse, Andrew N. Carr, Jan Leike, Josh Achiam, Vedant Misra, Evan Morikawa, Alec Radford, Matthew Knight, Miles Brundage, Mira Murati, Katie Mayer, Peter Welinder, Bob McGrew, Dario Amodei, Sam McCandlish, Ilya Sutskever, Wojciech Zaremba2021-07-07capabilitiessafety+31 articleView β†’
πŸ”— WebEvaluating Truthfulness: Benchmarking LLM Accuracyβ€”Noneβ€”β€”capabilitiesevaluation+1β€”View β†’
πŸ”— WebEvaluation methodologyMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluation1 articleView β†’
πŸ”— WebEvaluation MethodologyMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationmesa-optimization+32 articlesView β†’
πŸ”— Webevaluations.metr.orgβ€”Noneβ€”β€”evaluationevaluations+21 articleView β†’
πŸ”— Webevaluations.metr.orgβ€”Noneβ€”β€”evaluationevaluations+21 articleView β†’
✏️ Blog PostEvan HubingerLessWrongβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperEvans et al. (2021)arXivβ˜…β˜…β˜…β˜†β˜†SummaryRandall Balestriero, Jerome Pesenti, Yann LeCun2021-10-18capabilitiestraining+31 articleView β†’
πŸ“„ PaperEvidence from NatureNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebEvonetix Evaleoβ€”Noneβ€”β€”evaluationbiosecurity+21 articleView β†’
πŸ”— WebExecutive Order 14110RAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ›οΈ GovernmentExecutive Order 14110β€”Noneβ€”β€”β€”β€”View β†’
πŸ›οΈ GovernmentExecutive Order 14110β€”Noneβ€”β€”compute-thresholdsgovernance+11 articleView β†’
πŸ”— WebExecutive Order 14110 on AIβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentExecutive Order 14179β€”Noneβ€”β€”compute-thresholdsgovernance+11 articleView β†’
πŸ“„ PaperExecutive order blockedScienceβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ›οΈ GovernmentExecutive Order on AIWhite Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebExistential Risk Survey Results (EA Forum)EA Forumβ˜…β˜…β˜…β˜†β˜†MetadataRobBensinger2021-06-01x-risk1 articleView β†’
πŸ”— WebExpanded controlsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebExpert analysisAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”resource-allocationresearch-priorities+11 articleView β†’
πŸ”— WebExplainable and transparent artificial intelligence for public policymakingCambridge University Pressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”governanceai-bias+21 articleView β†’
πŸ›οΈ GovernmentExport control measures implemented in October 2022Bureau of Industry and Securityβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ›οΈ GovernmentExport controlsBureau of Industry and Securityβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorgame-theory+11 articleView β†’
πŸ›οΈ GovernmentExport controlsβ€”Noneβ€”β€”risk-factorgame-theory+11 articleView β†’
πŸ›οΈ GovernmentExport controlsBureau of Industry and Securityβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-forecastingcompute-trends+11 articleView β†’
πŸ›οΈ Governmentexport controls on advanced semiconductorsBureau of Industry and Securityβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ›οΈ GovernmentExport controls on advanced semiconductorsBureau of Industry and Securityβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebExtensive lobbyingReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorgame-theory+11 articleView β†’
πŸ”— WebExtracting Concepts from GPT-4OpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llm1 articleView β†’
πŸ”— WebEY - Major AI deal lifts Q1 2025 VC investment

EY provides insights into the current venture capital landscape, discussing investment challenges, market volatility, and potential opportunities for founders.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentFAA Human Factors Divisionβ€”Noneβ€”β€”automationhuman-factors+11 articleView β†’
πŸ›οΈ GovernmentFAA studiesβ€”Noneβ€”β€”automationhuman-factors+11 articleView β†’
πŸ”— WebFacebookβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebFacebook Oversight Board Reportsβ€”Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ”— WebFacebook's 2024 Coordinated Inauthentic Behavior Reportβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebFacebook's algorithm changes have historically affected global political discourseβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebFaceSwap benchmarksGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”capabilitiesevaluation+32 articlesView β†’
πŸ”— Webfact sheet statedβ€”Noneβ€”β€”compute-thresholdsgovernance+11 articleView β†’
πŸ”— WebFairlearn (Microsoft)β€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebFakespot analysis

Mozilla announced the shutdown of its Pocket and Fakespot products by July 2025, redirecting efforts towards enhancing the Firefox browser and developing new internet tools.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebFAR AIβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFAR AIβ€”Noneβ€”β€”adversarial-robustnessml-safety+11 articleView β†’
πŸ“– ReferenceFar-UVCWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ“„ PaperFar-UVC light: A new tool to control the spread of airborne-mediated microbial diseasesNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebFar-UVC researchβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebFarid: Digital image forensics

Hany Farid is a computer science professor specializing in digital forensics, image analysis, and detecting media manipulation. His research focuses on developing computational techniques to identify fake photos, videos, and AI-generated content.

β€”Fullβ€”β€”computeepistemic+32 articlesView β†’
πŸ”— WebFasken Comparative Analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFAT* Conferenceβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— Webfatal 2018 Uber self-driving car accident in Arizonaβ€”Noneβ€”β€”robustnessgeneralization+11 articleView β†’
πŸ›οΈ GovernmentFBI IC3β€”Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ›οΈ GovernmentFBI IC3 Reportsβ€”Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ›οΈ GovernmentFBI Internet Crime Reportβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentFBI Internet Crime Reportβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebFCA Guidance (UK)β€”Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebFCC Net Neutrality commentsThe New York Timesβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ›οΈ GovernmentFDA AI/ML Guidanceβ€”Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ›οΈ GovernmentFDA FAERS Databaseβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentFederal Communications Commission AI Guidelinesβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ›οΈ GovernmentFederal Register: Framework for AI Diffusion

The Bureau of Industry and Security (BIS) introduces new regulations controlling the export of advanced AI model weights and computing integrated circuits. The framework aims to balance national security concerns with enabling responsible global AI development.

β€”Fullβ€”β€”interpretabilitygovernance+12 articlesView β†’
πŸ›οΈ GovernmentFederal Reserveβ€”Noneβ€”β€”monitoringearly-warning+11 articleView β†’
πŸ”— Webfederated learningβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— Webfew manufacturersβ€”Noneβ€”β€”risk-factorgame-theory+11 articleView β†’
πŸ”— Webfewer than 20 organizations worldwideβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebFHI expert elicitationFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interventionseffectiveness+34 articlesView β†’
πŸ”— WebFHI publication guidelinesFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factordiffusion+11 articleView β†’
πŸ”— Webfiled an SEC complaintβ€”Noneβ€”β€”frontier-labssafety-culture+12 articlesView β†’
πŸ”— Webfinal reportβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebFinal Round AI

A comprehensive analysis of AI's immediate impact on job markets, highlighting widespread workforce reductions and the accelerating pace of job automation across multiple sectors.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— Webfinal version significantly narrowed its scopeβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFinancial Timesβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFinancial Times reportedβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFinancial Times: AI Competitionβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperFinn et al. (2017)arXivβ˜…β˜…β˜…β˜†β˜†SummaryChelsea Finn, Pieter Abbeel, Sergey Levine2017-03-09governancecapabilities+31 articleView β†’
πŸ”— WebFinnish Institute of International Affairsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFireEye Mandiantβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ›οΈ GovernmentFirst AISIC plenary meetingNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”governancegovernment-oversight+11 articleView β†’
πŸ”— WebFirst Amendment challengesβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebFirst autonomous coding agentβ€”Noneβ€”β€”tool-useagentic+11 articleView β†’
πŸ”— Webfirst batch of 30 providersCNBCβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”regulationchina+11 articleView β†’
πŸ”— Webfirst binding international AI treatyβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— Webfirst documented AI-orchestrated cyberattackAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebFirst Draft

First Draft developed comprehensive resources and research on understanding and addressing information disorder across six key categories. Their materials are available under a Creative Commons license.

β€”Fullβ€”β€”historical-evidencearchives+32 articlesView β†’
πŸ”— Webfirst fundraiser in six yearsMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”research-agendasalignment+11 articleView β†’
πŸ”— WebFirst intergovernmental AI dialogueβ€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— Webfirst meeting of the International NetworkEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebFive tech companies control over 80%β€”Noneβ€”β€”x-riskirreversibility+32 articlesView β†’
πŸ”— WebFLI AI Safety Index 2024Future of Life Instituteβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebFLI AI Safety Index Summer 2025Future of Life Institute

The FLI AI Safety Index Summer 2025 assesses leading AI companies' safety efforts, finding widespread inadequacies in risk management and existential safety planning. Anthropic leads with a C+ grade, while most companies score poorly across critical safety domains.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”safetyx-risk+39 articlesView β†’
πŸ”— WebFLI open letterFuture of Life Instituteβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”chatgptgpt-4+13 articlesView β†’
πŸ”— Webfootnote 17 problemMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFor Better Scienceβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebForbesβ€”Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
✏️ Blog PostForecasting AI Futures: AGI Insights from Prediction Marketsβ€”Noneβ€”β€”agi1 articleView β†’
πŸ”— WebForecasting analysis by Metaculus aggregatesMetaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ”— WebForecasting Research Institute

A research organization focused on advancing forecasting science through innovative methodologies and experimental approaches. They work with policymakers and nonprofits to develop practical prediction tools.

β€”Fullβ€”β€”governanceforecasting+21 articleView β†’
πŸ”— WebForecasting Research Instituteβ€”Noneβ€”β€”power-seekingself-preservation+11 articleView β†’
πŸ”— WebForeign Affairs: The Perilous Coming Age of AI Warfareβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebForesight Instituteβ€”Noneβ€”β€”interpretabilityscalable-oversight+11 articleView β†’
πŸ”— WebForethought Foundation's analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFortune AI training costsFortune

Research shows AI training costs are dramatically increasing, with models potentially costing billions of dollars and computational requirements doubling every six months. The trend raises questions about sustainability and future AI development.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”training1 articleView β†’
πŸ”— WebFortune AsiaFortuneβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFortune Business Insightsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFortune: AI and High-Frequency TradingFortune

The article explores the evolution of AI and algorithmic trading, examining its benefits and potential risks to financial markets. It highlights how high-frequency trading can create market instability and warns about potential challenges with generative AI trading tools.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”β€”β€”View β†’
πŸ”— WebFortune: Anthropic makes a breakthrough in opening AI's 'black box'Fortuneβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebFortune: Google DeepMind 145-page paper predicts AGI by 2030 (Apr 2025)Fortuneβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agi1 articleView β†’
πŸ”— WebFour Things to Know About China's New AI Rules in 2024MIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”regulationchina+11 articleView β†’
πŸ›οΈ GovernmentFourth Progress ReportUK AI Safety Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancegovernment-ai-safety+11 articleView β†’
πŸ›οΈ GovernmentFramework for Nucleic Acid Synthesis Screeningβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebFrance's AI Action SummitCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalgovernance+11 articleView β†’
πŸ”— WebFrance24, 2023β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFranΓ§ois Cholletβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFraud detection guidelinesβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebFreedom HouseFreedom Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”authoritarianismhuman-rights+33 articlesView β†’
πŸ”— WebFreedom HouseFreedom Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ”— WebFreedom HouseFreedom Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ”— WebFreedom House ReportsFreedom Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebFreedom House trackingFreedom Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebFreedom House's 2024 assessmentFreedom Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebFreedom on the Net 2025Freedom Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFreedom Online Coalitionβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebFrom Predicting Dissent to Programming Power: AI-Driven Authoritarian Governanceβ€”Noneβ€”β€”governancex-risk+11 articleView β†’
πŸ›οΈ GovernmentFrontier AI capability evaluationUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitiesevaluation1 articleView β†’
πŸ”— WebFrontier AI Safety Commitmentsβ€”Noneβ€”β€”safety3 articlesView β†’
πŸ”— WebFrontier AI Safety CommitmentsCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyregulation+21 articleView β†’
πŸ”— WebFrontier Model Forumβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebFrontier Model Forumβ€”Noneβ€”β€”risk-factorgame-theory+11 articleView β†’
πŸ”— WebFrontier Model Forumβ€”Noneβ€”β€”β€”3 articlesView β†’
πŸ”— WebFrontier Model Forum'sβ€”Noneβ€”β€”game-theorygovernance+32 articlesView β†’
πŸ”— WebFrontier Models are Capable of In-Context SchemingApollo Researchβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deceptionself-awareness+37 articlesView β†’
πŸ”— WebFrontier SafetyGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetygemini+21 articleView β†’
πŸ”— WebFrontier Safety Framework v2Google DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetytool-use+21 articleView β†’
πŸ”— WebFrontierMath benchmarkEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitiesevaluation1 articleView β†’
πŸ”— WebFrontiersβ€”Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ”— WebFrontiers in AIβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebFrontiers in Energy Researchβ€”Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ”— WebFrontiers in Political Science researchβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebFT AI Coverageβ€”Noneβ€”β€”constitutional-airesponsible-scaling+11 articleView β†’
πŸ›οΈ GovernmentFTCFederal Trade Commissionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ›οΈ GovernmentFTC Dark Patterns enforcementFederal Trade Commissionβ˜…β˜…β˜…β˜…β˜†Metadataβ€”2021ai-ethicspersuasion+11 articleView β†’
πŸ›οΈ GovernmentFTC enforcement actionsFederal Trade Commission

The Federal Trade Commission (FTC) enforces over 70 laws to prevent fraud, deception, and anticompetitive business practices. Its mission is to protect consumers and maintain fair market competition.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”deceptionβ€”View β†’
πŸ›οΈ GovernmentFTC fake review enforcementFederal Trade Commissionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ›οΈ GovernmentFTC reportsFederal Trade Commissionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”timelinecapability+11 articleView β†’
πŸ›οΈ GovernmentFTC: Crackdown on Deceptive AI ClaimsFederal Trade Commission

The Federal Trade Commission initiated a law enforcement sweep targeting companies using AI technology to engage in fraudulent business practices. The actions focus on preventing deceptive claims and protecting consumers from misleading AI-powered services.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”deceptionβ€”View β†’
πŸ›οΈ GovernmentFTC's August 2024 final ruleFederal Trade Commissionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ›οΈ GovernmentFTC's investigationFederal Trade Commissionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+32 articlesView β†’
πŸ”— Webfull comparison between public and Anthropic constitutionsβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebFull Fact

Full Fact is a non-profit fact-checking organization that monitors public discourse, investigates false claims, and promotes media literacy. They use AI tools to identify and combat misinformation across various domains.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— Webfunctional decision theoryMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFunSearchGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFuture Fundβ€”Noneβ€”β€”cost-effectivenessresearch-priorities+11 articleView β†’
πŸ”— WebFuture of Humanity InstituteFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorgame-theory+11 articleView β†’
πŸ”— WebFuture of Humanity Institute (2019)Future of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”escalationconflict+32 articlesView β†’
πŸ”— WebFuture of Humanity Institute researchFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFuture of Humanity Institute surveysFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFuture of Life InstituteFuture of Life Institute

The Future of Life Institute works to guide transformative technologies like AI towards beneficial outcomes and away from large-scale risks. They engage in policy advocacy, research, education, and grantmaking to promote safe and responsible technological development.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”governancesafety+32 articlesView β†’
πŸ”— WebFuture of Life InstituteFuture of Life Institute

The Vitalik Buterin PhD Fellowship supports students researching ways to reduce existential risks from advanced AI technologies. Fellows receive funding, research support, and networking opportunities.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”x-riskβ€”View β†’
πŸ”— WebFuture of Life InstituteTIMEβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebFuture of Life Institute notesFuture of Life Instituteβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebFuture of Life Institute: AI Safety Index 2024Future of Life Institute

The Future of Life Institute's AI Safety Index 2024 evaluates six leading AI companies across 42 safety indicators, highlighting major concerns about risk management and potential AI threats.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”safetyevaluation+35 articlesView β†’
πŸ”— WebFuture of Life Institute: Existential Risk PodcastsFuture of Life Instituteβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”x-riskbiosecurity+11 articleView β†’
πŸ”— WebFuture of Life Institute's research with Mithril SecurityFuture of Life Instituteβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”cybersecurity1 articleView β†’
πŸ“„ PaperFutureSearch paperarXiv

The paper investigates the spin conductivity distribution in disordered quantum spin chains, demonstrating that while the average conductivity suggests metallicity, the typical conductivity indicates an insulating state.

β˜…β˜…β˜…β˜†β˜†FullFaria, L. F. C., Quito, Victor L., Getelina, JoΓ£o C., Hoyos, JosΓ© A., Miranda, E.2023forecastingprediction-markets+11 articleView β†’
πŸ”— WebFuturism: Google AI Boss Says AI Is an Existential Threatβ€”Noneβ€”β€”x-risk1 articleView β†’
πŸ”— WebG7 Hiroshima AI Processβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebG7's Hiroshima AI ProcessEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebG7/G20 coordination working groupsβ€”Noneβ€”β€”risk-factorgame-theory+11 articleView β†’
πŸ“„ PaperGAIAarXivβ˜…β˜…β˜…β˜†β˜†SummaryGrΓ©goire Mialon, ClΓ©mentine Fourrier, Craig Swift, Thomas Wolf, Yann LeCun, Thomas Scialom2023-11-21capabilitiesevaluation+31 articleView β†’
πŸ”— WebGallupGallup

A 2025 Gallup survey shows Americans increasingly neutral about AI's impact, with 31% trusting businesses to use AI responsibly. Concerns persist about job market disruption.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”economic1 articleView β†’
πŸ”— WebGallupGallupβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”monitoringearly-warning+11 articleView β†’
πŸ”— WebGallupGallupβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebGallup (2023)Gallupβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”information-overloadmedia-literacy+11 articleView β†’
πŸ”— WebGallup AI Safety PollGallup

A national Gallup survey shows 80% of Americans prioritize AI safety rules over rapid development, with broad support for government oversight and independent testing of AI technologies.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”safetyevaluationβ€”View β†’
πŸ”— WebGallup Trust SurveysGallupβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”information-overloadmedia-literacy+11 articleView β†’
πŸ”— WebGallup: 32% trustGallupβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebGallup: Confidence in InstitutionsGallup

A survey assessing public trust and confidence levels across different institutions in American society. Examines perceptions of key organizations and sectors.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”epistemiccascade+33 articlesView β†’
πŸ”— WebGallup: Historic lowsGallupβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebGallup's October 2025 surveyGallupβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”truthepistemology+12 articlesView β†’
πŸ“„ PaperGame-theoretic analysisarXivβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGame-theoretic modeling by Georgetown's Center for Security and Emerging TechnologyCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecuritygame-theory+21 articleView β†’
πŸ“„ PaperGaming RLHF evaluationarXivβ˜…β˜…β˜…β˜†β˜†SummaryRichard Ngo, Lawrence Chan, SΓΆren Mindermann2022-08-30alignmentcapabilities+34 articlesView β†’
πŸ”— WebGans (2024): Market Power in Artificial Intelligenceβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentGAO-24-107292β€”Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ›οΈ GovernmentGAO: AI Agencies Implementing Management Requirementsβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebGartnerβ€”Noneβ€”β€”tool-useagentic+11 articleView β†’
πŸ”— WebGartner predictsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGartner predictsβ€”Noneβ€”β€”tool-useagentic+11 articleView β†’
πŸ”— WebGartner predicts 40%+ agentic AI projects will be cancelled by 2027β€”Noneβ€”β€”tool-useagentic+11 articleView β†’
πŸ”— WebGartner/DemandSageβ€”Noneβ€”β€”labor-marketsautomation+11 articleView β†’
πŸ”— WebGary Marcus: Deep Learning Alone Won't Get Us to AGIβ€”Noneβ€”β€”agi1 articleView β†’
✏️ Blog PostGary Marcus's Substack

Gary Marcus's Substack offers expert analysis and commentary on artificial intelligence, focusing on responsible AI development and potential risks.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebGCGβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentGCHQβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebGeddes-Wright-Frantz Autocratic Regimes datasetCambridge University Pressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”authoritarianismstability+11 articleView β†’
πŸ”— WebGemini 1.0 UltraGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llmregulation+35 articlesView β†’
πŸ”— WebGemini 2.0 FlashGoogle AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llm2 articlesView β†’
πŸ“„ PaperGemini ReportarXivβ˜…β˜…β˜…β˜†β˜†SummaryGemini Team, Rohan Anil, Sebastian Borgeaud, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M. Dai, Anja Hauth, Katie Millican, David Silver, Melvin Johnson, Ioannis Antonoglou, Julian Schrittwieser, Amelia Glaese, Jilin Chen, Emily Pitler, Timothy Lillicrap, Angeliki Lazaridou, Orhan Firat, James Molloy, Michael Isard, Paul R. Barham, Tom Hennigan, Benjamin Lee, Fabio Viola, Malcolm Reynolds, Yuanzhong Xu, Ryan Doherty, Eli Collins, Clemens Meyer, Eliza Rutherford, Erica Moreira, Kareem Ayoub, Megha Goel, Jack Krawczyk, Cosmo Du, Ed Chi, Heng-Tze Cheng, Eric Ni, Purvi Shah, Patrick Kane, Betty Chan, Manaal Faruqui, Aliaksei Severyn, Hanzhao Lin, YaGuang Li, Yong Cheng, Abe Ittycheriah, Mahdis Mahdieh, Mia Chen, Pei Sun, Dustin Tran, Sumit Bagri, Balaji Lakshminarayanan, Jeremiah Liu, Andras Orban, Fabian GΓΌra, Hao Zhou, Xinying Song, Aurelien Boffy, Harish Ganapathy, Steven Zheng, HyunJeong Choe, Ágoston Weisz, Tao Zhu, Yifeng Lu, Siddharth Gopal, Jarrod Kahn, Maciej Kula, Jeff Pitman, Rushin Shah, Emanuel Taropa, Majd Al Merey, Martin Baeuml, Zhifeng Chen, Laurent El Shafey, Yujing Zhang, Olcan Sercinoglu, George Tucker, Enrique Piqueras, Maxim Krikun, Iain Barr, Nikolay Savinov, Ivo Danihelka, Becca Roelofs, AnaΓ―s White, Anders Andreassen, Tamara von Glehn, Lakshman Yagati, Mehran Kazemi, Lucas Gonzalez, Misha Khalman, Jakub Sygnowski, Alexandre Frechette, Charlotte Smith, Laura Culp, Lev Proleev, Yi Luan, Xi Chen, James Lottes, Nathan Schucher, Federico Lebron, Alban Rrustemi, Natalie Clay, Phil Crone, Tomas Kocisky, Jeffrey Zhao, Bartek Perz, Dian Yu, Heidi Howard, Adam Bloniarz, Jack W. Rae, Han Lu, Laurent Sifre, Marcello Maggioni, Fred Alcober, Dan Garrette, Megan Barnes, Shantanu Thakoor, Jacob Austin, Gabriel Barth-Maron, William Wong, Rishabh Joshi, Rahma Chaabouni, Deeni Fatiha, Arun Ahuja, Gaurav Singh Tomar, Evan Senter, Martin Chadwick, Ilya Kornakov, Nithya Attaluri, IΓ±aki Iturrate, Ruibo Liu, Yunxuan Li, Sarah Cogan, Jeremy Chen, Chao Jia, Chenjie Gu, Qiao Zhang, Jordan Grimstad, Ale Jakse Hartman, Xavier Garcia, Thanumalayan Sankaranarayana Pillai, Jacob Devlin, Michael Laskin, Diego de Las Casas, Dasha Valter, Connie Tao, Lorenzo Blanco, AdriΓ  PuigdomΓ¨nech Badia, David Reitter, Mianna Chen, Jenny Brennan, Clara Rivera, Sergey Brin, Shariq Iqbal, Gabriela Surita, Jane Labanowski, Abhi Rao, Stephanie Winkler, Emilio Parisotto, Yiming Gu, Kate Olszewska, Ravi Addanki, Antoine Miech, Annie Louis, Denis Teplyashin, Geoff Brown, Elliot Catt, Jan Balaguer, Jackie Xiang, Pidong Wang, Zoe Ashwood, Anton Briukhov, Albert Webson, Sanjay Ganapathy, Smit Sanghavi, Ajay Kannan, Ming-Wei Chang, Axel Stjerngren, Josip Djolonga, Yuting Sun, Ankur Bapna, Matthew Aitchison, Pedram Pejman, Henryk Michalewski, Tianhe Yu, Cindy Wang, Juliette Love, Junwhan Ahn, Dawn Bloxwich, Kehang Han, Peter Humphreys, Thibault Sellam, James Bradbury, Varun Godbole, Sina Samangooei, Bogdan Damoc, Alex Kaskasoli, SΓ©bastien M. R. Arnold, Vijay Vasudevan, Shubham Agrawal, Jason Riesa, Dmitry Lepikhin, Richard Tanburn, Srivatsan Srinivasan, Hyeontaek Lim, Sarah Hodkinson, Pranav Shyam, Johan Ferret, Steven Hand, Ankush Garg, Tom Le Paine, Jian Li, Yujia Li, Minh Giang, Alexander Neitz, Zaheer Abbas, Sarah York, Machel Reid, Elizabeth Cole, Aakanksha Chowdhery, Dipanjan Das, Dominika RogoziΕ„ska, Vitaliy Nikolaev, Pablo Sprechmann, Zachary Nado, Lukas Zilka, Flavien Prost, Luheng He, Marianne Monteiro, Gaurav Mishra, Chris Welty, Josh Newlan, Dawei Jia, Miltiadis Allamanis, Clara Huiyi Hu, Raoul de Liedekerke, Justin Gilmer, Carl Saroufim, Shruti Rijhwani, Shaobo Hou, Disha Shrivastava, Anirudh Baddepudi, Alex Goldin, Adnan Ozturel, Albin Cassirer, Yunhan Xu, Daniel Sohn, Devendra Sachan, Reinald Kim Amplayo, Craig Swanson, Dessie Petrova, Shashi Narayan, Arthur Guez, Siddhartha Brahma, Jessica Landon, Miteyan Patel, Ruizhe Zhao, Kevin Villela, Luyu Wang, Wenhao Jia, Matthew Rahtz, Mai GimΓ©nez, Legg Yeung, James Keeling, Petko Georgiev, Diana Mincu, Boxi Wu, Salem Haykal, Rachel Saputro, Kiran Vodrahalli, James Qin, Zeynep Cankara, Abhanshu Sharma, Nick Fernando, Will Hawkins, Behnam Neyshabur, Solomon Kim, Adrian Hutter, Priyanka Agrawal, Alex Castro-Ros, George van den Driessche, Tao Wang, Fan Yang, Shuo-yiin Chang, Paul Komarek, Ross McIlroy, Mario LučiΔ‡, Guodong Zhang, Wael Farhan, Michael Sharman, Paul Natsev, Paul Michel, Yamini Bansal, Siyuan Qiao, Kris Cao, Siamak Shakeri, Christina Butterfield, Justin Chung, Paul Kishan Rubenstein, Shivani Agrawal, Arthur Mensch, Kedar Soparkar, Karel Lenc, Timothy Chung, Aedan Pope, Loren Maggiore, Jackie Kay, Priya Jhakra, Shibo Wang, Joshua Maynez, Mary Phuong, Taylor Tobin, Andrea Tacchetti, Maja Trebacz, Kevin Robinson, Yash Katariya, Sebastian Riedel, Paige Bailey, Kefan Xiao, Nimesh Ghelani, Lora Aroyo, Ambrose Slone, Neil Houlsby, Xuehan Xiong, Zhen Yang, Elena Gribovskaya, Jonas Adler, Mateo Wirth, Lisa Lee, Music Li, Thais Kagohara, Jay Pavagadhi, Sophie Bridgers, Anna Bortsova, Sanjay Ghemawat, Zafarali Ahmed, Tianqi Liu, Richard Powell, Vijay Bolina, Mariko Iinuma, Polina Zablotskaia, James Besley, Da-Woon Chung, Timothy Dozat, Ramona Comanescu, Xiance Si, Jeremy Greer, Guolong Su, Martin Polacek, RaphaΓ«l Lopez Kaufman, Simon Tokumine, Hexiang Hu, Elena Buchatskaya, Yingjie Miao, Mohamed Elhawaty, Aditya Siddhant, Nenad Tomasev, Jinwei Xing, Christina Greer, Helen Miller, Shereen Ashraf, Aurko Roy, Zizhao Zhang, Ada Ma, Angelos Filos, Milos Besta, Rory Blevins, Ted Klimenko, Chih-Kuan Yeh, Soravit Changpinyo, Jiaqi Mu, Oscar Chang, Mantas Pajarskas, Carrie Muir, Vered Cohen, Charline Le Lan, Krishna Haridasan, Amit Marathe, Steven Hansen, Sholto Douglas, Rajkumar Samuel, Mingqiu Wang, Sophia Austin, Chang Lan, Jiepu Jiang, Justin Chiu, Jaime Alonso Lorenzo, Lars Lowe SjΓΆsund, SΓ©bastien Cevey, Zach Gleicher, Thi Avrahami, Anudhyan Boral, Hansa Srinivasan, Vittorio Selo, Rhys May, Konstantinos Aisopos, LΓ©onard Hussenot, Livio Baldini Soares, Kate Baumli, Michael B. Chang, AdriΓ  Recasens, Ben Caine, Alexander Pritzel, Filip Pavetic, Fabio Pardo, Anita Gergely, Justin Frye, Vinay Ramasesh, Dan Horgan, Kartikeya Badola, Nora Kassner, Subhrajit Roy, Ethan Dyer, VΓ­ctor Campos Campos, Alex Tomala, Yunhao Tang, Dalia El Badawy, Elspeth White, Basil Mustafa, Oran Lang, Abhishek Jindal, Sharad Vikram, Zhitao Gong, Sergi Caelles, Ross Hemsley, Gregory Thornton, Fangxiaoyu Feng, Wojciech Stokowiec, Ce Zheng, Phoebe Thacker, Γ‡ağlar ÜnlΓΌ, Zhishuai Zhang, Mohammad Saleh, James Svensson, Max Bileschi, Piyush Patil, Ankesh Anand, Roman Ring, Katerina Tsihlas, Arpi Vezer, Marco Selvi, Toby Shevlane, Mikel Rodriguez, Tom Kwiatkowski, Samira Daruki, Keran Rong, Allan Dafoe, Nicholas FitzGerald, Keren Gu-Lemberg, Mina Khan, Lisa Anne Hendricks, Marie Pellat, Vladimir Feinberg, James Cobon-Kerr, Tara Sainath, Maribeth Rauh, Sayed Hadi Hashemi, Richard Ives, Yana Hasson, Eric Noland, Yuan Cao, Nathan Byrd, Le Hou, Qingze Wang, Thibault Sottiaux, Michela Paganini, Jean-Baptiste Lespiau, Alexandre Moufarek, Samer Hassan, Kaushik Shivakumar, Joost van Amersfoort, Amol Mandhane, Pratik Joshi, Anirudh Goyal, Matthew Tung, Andrew Brock, Hannah Sheahan, Vedant Misra, Cheng Li, Nemanja RakiΔ‡eviΔ‡, Mostafa Dehghani, Fangyu Liu, Sid Mittal, Junhyuk Oh, Seb Noury, Eren Sezener, Fantine Huot, Matthew Lamm, Nicola De Cao, Charlie Chen, Sidharth Mudgal, Romina Stella, Kevin Brooks, Gautam Vasudevan, Chenxi Liu, Mainak Chain, Nivedita Melinkeri, Aaron Cohen, Venus Wang, Kristie Seymore, Sergey Zubkov, Rahul Goel, Summer Yue, Sai Krishnakumaran, Brian Albert, Nate Hurley, Motoki Sano, Anhad Mohananey, Jonah Joughin, Egor Filonov, Tomasz KΔ™pa, Yomna Eldawy, Jiawern Lim, Rahul Rishi, Shirin Badiezadegan, Taylor Bos, Jerry Chang, Sanil Jain, Sri Gayatri Sundara Padmanabhan, Subha Puttagunta, Kalpesh Krishna, Leslie Baker, Norbert Kalb, Vamsi Bedapudi, Adam Kurzrok, Shuntong Lei, Anthony Yu, Oren Litvin, Xiang Zhou, Zhichun Wu, Sam Sobell, Andrea Siciliano, Alan Papir, Robby Neale, Jonas Bragagnolo, Tej Toor, Tina Chen, Valentin Anklin, Feiran Wang, Richie Feng, Milad Gholami, Kevin Ling, Lijuan Liu, Jules Walter, Hamid Moghaddam, Arun Kishore, Jakub Adamek, Tyler Mercado, Jonathan Mallinson, Siddhinita Wandekar, Stephen Cagle, Eran Ofek, Guillermo Garrido, Clemens Lombriser, Maksim Mukha, Botu Sun, Hafeezul Rahman Mohammad, Josip Matak, Yadi Qian, Vikas Peswani, Pawel Janus, Quan Yuan, Leif Schelin, Oana David, Ankur Garg, Yifan He, Oleksii Duzhyi, Anton Γ„lgmyr, TimothΓ©e Lottaz, Qi Li, Vikas Yadav, Luyao Xu, Alex Chinien, Rakesh Shivanna, Aleksandr Chuklin, Josie Li, Carrie Spadine, Travis Wolfe, Kareem Mohamed, Subhabrata Das, Zihang Dai, Kyle He, Daniel von Dincklage, Shyam Upadhyay, Akanksha Maurya, Luyan Chi, Sebastian Krause, Khalid Salama, Pam G Rabinovitch, Pavan Kumar Reddy M, Aarush Selvan, Mikhail Dektiarev, Golnaz Ghiasi, Erdem Guven, Himanshu Gupta, Boyi Liu, Deepak Sharma, Idan Heimlich Shtacher, Shachi Paul, Oscar Akerlund, FranΓ§ois-Xavier Aubet, Terry Huang, Chen Zhu, Eric Zhu, Elico Teixeira, Matthew Fritze, Francesco Bertolini, Liana-Eleonora Marinescu, Martin BΓΆlle, Dominik Paulus, Khyatti Gupta, Tejasi Latkar, Max Chang, Jason Sanders, Roopa Wilson, Xuewei Wu, Yi-Xuan Tan, Lam Nguyen Thiet, Tulsee Doshi, Sid Lall, Swaroop Mishra, Wanming Chen, Thang Luong, Seth Benjamin, Jasmine Lee, Ewa Andrejczuk, Dominik Rabiej, Vipul Ranjan, Krzysztof Styrc, Pengcheng Yin, Jon Simon, Malcolm Rose Harriott, Mudit Bansal, Alexei Robsky, Geoff Bacon, David Greene, Daniil Mirylenka, Chen Zhou, Obaid Sarvana, Abhimanyu Goyal, Samuel Andermatt, Patrick Siegler, Ben Horn, Assaf Israel, Francesco Pongetti, Chih-Wei "Louis" Chen, Marco Selvatici, Pedro Silva, Kathie Wang, Jackson Tolins, Kelvin Guu, Roey Yogev, Xiaochen Cai, Alessandro Agostini, Maulik Shah, Hung Nguyen, Noah Γ“ Donnaile, SΓ©bastien Pereira, Linda Friso, Adam Stambler, Adam Kurzrok, Chenkai Kuang, Yan Romanikhin, Mark Geller, ZJ Yan, Kane Jang, Cheng-Chun Lee, Wojciech Fica, Eric Malmi, Qijun Tan, Dan Banica, Daniel Balle, Ryan Pham, Yanping Huang, Diana Avram, Hongzhi Shi, Jasjot Singh, Chris Hidey, Niharika Ahuja, Pranab Saxena, Dan Dooley, Srividya Pranavi Potharaju, Eileen O'Neill, Anand Gokulchandran, Ryan Foley, Kai Zhao, Mike Dusenberry, Yuan Liu, Pulkit Mehta, Ragha Kotikalapudi, Chalence Safranek-Shrader, Andrew Goodman, Joshua Kessinger, Eran Globen, Prateek Kolhar, Chris Gorgolewski, Ali Ibrahim, Yang Song, Ali Eichenbaum, Thomas Brovelli, Sahitya Potluri, Preethi Lahoti, Cip Baetu, Ali Ghorbani, Charles Chen, Andy Crawford, Shalini Pal, Mukund Sridhar, Petru Gurita, Asier Mujika, Igor Petrovski, Pierre-Louis Cedoz, Chenmei Li, Shiyuan Chen, NiccolΓ² Dal Santo, Siddharth Goyal, Jitesh Punjabi, Karthik Kappaganthu, Chester Kwak, Pallavi LV, Sarmishta Velury, Himadri Choudhury, Jamie Hall, Premal Shah, Ricardo Figueira, Matt Thomas, Minjie Lu, Ting Zhou, Chintu Kumar, Thomas Jurdi, Sharat Chikkerur, Yenai Ma, Adams Yu, Soo Kwak, Victor Γ„hdel, Sujeevan Rajayogam, Travis Choma, Fei Liu, Aditya Barua, Colin Ji, Ji Ho Park, Vincent Hellendoorn, Alex Bailey, Taylan Bilal, Huanjie Zhou, Mehrdad Khatir, Charles Sutton, Wojciech Rzadkowski, Fiona Macintosh, Roopali Vij, Konstantin Shagin, Paul Medina, Chen Liang, Jinjing Zhou, Pararth Shah, Yingying Bi, Attila Dankovics, Shipra Banga, Sabine Lehmann, Marissa Bredesen, Zifan Lin, John Eric Hoffmann, Jonathan Lai, Raynald Chung, Kai Yang, Nihal Balani, Arthur BraΕΎinskas, Andrei Sozanschi, Matthew Hayes, HΓ©ctor FernΓ‘ndez Alcalde, Peter Makarov, Will Chen, Antonio Stella, Liselotte Snijders, Michael Mandl, Ante KΓ€rrman, PaweΕ‚ Nowak, Xinyi Wu, Alex Dyck, Krishnan Vaidyanathan, Raghavender R, Jessica Mallet, Mitch Rudominer, Eric Johnston, Sushil Mittal, Akhil Udathu, Janara Christensen, Vishal Verma, Zach Irving, Andreas Santucci, Gamaleldin Elsayed, Elnaz Davoodi, Marin Georgiev, Ian Tenney, Nan Hua, Geoffrey Cideron, Edouard Leurent, Mahmoud Alnahlawi, Ionut Georgescu, Nan Wei, Ivy Zheng, Dylan Scandinaro, Heinrich Jiang, Jasper Snoek, Mukund Sundararajan, Xuezhi Wang, Zack Ontiveros, Itay Karo, Jeremy Cole, Vinu Rajashekhar, Lara Tumeh, Eyal Ben-David, Rishub Jain, Jonathan Uesato, Romina Datta, Oskar Bunyan, Shimu Wu, John Zhang, Piotr Stanczyk, Ye Zhang, David Steiner, Subhajit Naskar, Michael Azzam, Matthew Johnson, Adam Paszke, Chung-Cheng Chiu, Jaume Sanchez Elias, Afroz Mohiuddin, Faizan Muhammad, Jin Miao, Andrew Lee, Nino Vieillard, Jane Park, Jiageng Zhang, Jeff Stanway, Drew Garmon, Abhijit Karmarkar, Zhe Dong, Jong Lee, Aviral Kumar, Luowei Zhou, Jonathan Evens, William Isaac, Geoffrey Irving, Edward Loper, Michael Fink, Isha Arkatkar, Nanxin Chen, Izhak Shafran, Ivan Petrychenko, Zhe Chen, Johnson Jia, Anselm Levskaya, Zhenkai Zhu, Peter Grabowski, Yu Mao, Alberto Magni, Kaisheng Yao, Javier Snaider, Norman Casagrande, Evan Palmer, Paul Suganthan, Alfonso CastaΓ±o, Irene Giannoumis, Wooyeol Kim, MikoΕ‚aj RybiΕ„ski, Ashwin Sreevatsa, Jennifer Prendki, David Soergel, Adrian Goedeckemeyer, Willi Gierke, Mohsen Jafari, Meenu Gaba, Jeremy Wiesner, Diana Gage Wright, Yawen Wei, Harsha Vashisht, Yana Kulizhskaya, Jay Hoover, Maigo Le, Lu Li, Chimezie Iwuanyanwu, Lu Liu, Kevin Ramirez, Andrey Khorlin, Albert Cui, Tian LIN, Marcus Wu, Ricardo Aguilar, Keith Pallo, Abhishek Chakladar, Ginger Perng, Elena Allica Abellan, Mingyang Zhang, Ishita Dasgupta, Nate Kushman, Ivo Penchev, Alena Repina, Xihui Wu, Tom van der Weide, Priya Ponnapalli, Caroline Kaplan, Jiri Simsa, Shuangfeng Li, Olivier Dousse, Fan Yang, Jeff Piper, Nathan Ie, Rama Pasumarthi, Nathan Lintz, Anitha Vijayakumar, Daniel Andor, Pedro Valenzuela, Minnie Lui, Cosmin Paduraru, Daiyi Peng, Katherine Lee, Shuyuan Zhang, Somer Greene, Duc Dung Nguyen, Paula Kurylowicz, Cassidy Hardin, Lucas Dixon, Lili Janzer, Kiam Choo, Ziqiang Feng, Biao Zhang, Achintya Singhal, Dayou Du, Dan McKinnon, Natasha Antropova, Tolga Bolukbasi, Orgad Keller, David Reid, Daniel Finchelstein, Maria Abi Raad, Remi Crocker, Peter Hawkins, Robert Dadashi, Colin Gaffney, Ken Franko, Anna Bulanova, RΓ©mi Leblond, Shirley Chung, Harry Askham, Luis C. Cobo, Kelvin Xu, Felix Fischer, Jun Xu, Christina Sorokin, Chris Alberti, Chu-Cheng Lin, Colin Evans, Alek Dimitriev, Hannah Forbes, Dylan Banarse, Zora Tung, Mark Omernick, Colton Bishop, Rachel Sterneck, Rohan Jain, Jiawei Xia, Ehsan Amid, Francesco Piccinno, Xingyu Wang, Praseem Banzal, Daniel J. Mankowitz, Alex Polozov, Victoria Krakovna, Sasha Brown, MohammadHossein Bateni, Dennis Duan, Vlad Firoiu, Meghana Thotakuri, Tom Natan, Matthieu Geist, Ser tan Girgin, Hui Li, Jiayu Ye, Ofir Roval, Reiko Tojo, Michael Kwong, James Lee-Thorp, Christopher Yew, Danila Sinopalnikov, Sabela Ramos, John Mellor, Abhishek Sharma, Kathy Wu, David Miller, Nicolas Sonnerat, Denis Vnukov, Rory Greig, Jennifer Beattie, Emily Caveness, Libin Bai, Julian Eisenschlos, Alex Korchemniy, Tomy Tsai, Mimi Jasarevic, Weize Kong, Phuong Dao, Zeyu Zheng, Frederick Liu, Fan Yang, Rui Zhu, Tian Huey Teh, Jason Sanmiya, Evgeny Gladchenko, Nejc Trdin, Daniel Toyama, Evan Rosen, Sasan Tavakkol, Linting Xue, Chen Elkind, Oliver Woodman, John Carpenter, George Papamakarios, Rupert Kemp, Sushant Kafle, Tanya Grunina, Rishika Sinha, Alice Talbert, Diane Wu, Denese Owusu-Afriyie, Cosmo Du, Chloe Thornton, Jordi Pont-Tuset, Pradyumna Narayana, Jing Li, Saaber Fatehi, John Wieting, Omar Ajmeri, Benigno Uria, Yeongil Ko, Laura Knight, AmΓ©lie HΓ©liou, Ning Niu, Shane Gu, Chenxi Pang, Yeqing Li, Nir Levine, Ariel Stolovich, Rebeca Santamaria-Fernandez, Sonam Goenka, Wenny Yustalim, Robin Strudel, Ali Elqursh, Charlie Deck, Hyo Lee, Zonglin Li, Kyle Levin, Raphael Hoffmann, Dan Holtmann-Rice, Olivier Bachem, Sho Arora, Christy Koh, Soheil Hassas Yeganeh, Siim PΓ΅der, Mukarram Tariq, Yanhua Sun, Lucian Ionita, Mojtaba Seyedhosseini, Pouya Tafti, Zhiyu Liu, Anmol Gulati, Jasmine Liu, Xinyu Ye, Bart Chrzaszcz, Lily Wang, Nikhil Sethi, Tianrun Li, Ben Brown, Shreya Singh, Wei Fan, Aaron Parisi, Joe Stanton, Vinod Koverkathu, Christopher A. Choquette-Choo, Yunjie Li, TJ Lu, Abe Ittycheriah, Prakash Shroff, Mani Varadarajan, Sanaz Bahargam, Rob Willoughby, David Gaddy, Guillaume Desjardins, Marco Cornero, Brona Robenek, Bhavishya Mittal, Ben Albrecht, Ashish Shenoy, Fedor Moiseev, Henrik Jacobsson, Alireza Ghaffarkhah, Morgane RiviΓ¨re, Alanna Walton, ClΓ©ment Crepy, Alicia Parrish, Zongwei Zhou, Clement Farabet, Carey Radebaugh, Praveen Srinivasan, Claudia van der Salm, Andreas Fidjeland, Salvatore Scellato, Eri Latorre-Chimoto, Hanna Klimczak-PluciΕ„ska, David Bridson, Dario de Cesare, Tom Hudson, Piermaria Mendolicchio, Lexi Walker, Alex Morris, Matthew Mauger, Alexey Guseynov, Alison Reid, Seth Odoom, Lucia Loher, Victor Cotruta, Madhavi Yenugula, Dominik Grewe, Anastasia Petrushkina, Tom Duerig, Antonio Sanchez, Steve Yadlowsky, Amy Shen, Amir Globerson, Lynette Webb, Sahil Dua, Dong Li, Surya Bhupatiraju, Dan Hurt, Haroon Qureshi, Ananth Agarwal, Tomer Shani, Matan Eyal, Anuj Khare, Shreyas Rammohan Belle, Lei Wang, Chetan Tekur, Mihir Sanjay Kale, Jinliang Wei, Ruoxin Sang, Brennan Saeta, Tyler Liechty, Yi Sun, Yao Zhao, Stephan Lee, Pandu Nayak, Doug Fritz, Manish Reddy Vuyyuru, John Aslanides, Nidhi Vyas, Martin Wicke, Xiao Ma, Evgenii Eltyshev, Nina Martin, Hardie Cate, James Manyika, Keyvan Amiri, Yelin Kim, Xi Xiong, Kai Kang, Florian Luisier, Nilesh Tripuraneni, David Madras, Mandy Guo, Austin Waters, Oliver Wang, Joshua Ainslie, Jason Baldridge, Han Zhang, Garima Pruthi, Jakob Bauer, Feng Yang, Riham Mansour, Jason Gelman, Yang Xu, George Polovets, Ji Liu, Honglong Cai, Warren Chen, XiangHai Sheng, Emily Xue, Sherjil Ozair, Christof Angermueller, Xiaowei Li, Anoop Sinha, Weiren Wang, Julia Wiesinger, Emmanouil Koukoumidis, Yuan Tian, Anand Iyer, Madhu Gurumurthy, Mark Goldenson, Parashar Shah, MK Blake, Hongkun Yu, Anthony Urbanowicz, Jennimaria Palomaki, Chrisantha Fernando, Ken Durden, Harsh Mehta, Nikola Momchev, Elahe Rahimtoroghi, Maria Georgaki, Amit Raul, Sebastian Ruder, Morgan Redshaw, Jinhyuk Lee, Denny Zhou, Komal Jalan, Dinghua Li, Blake Hechtman, Parker Schuh, Milad Nasr, Kieran Milan, Vladimir Mikulik, Juliana Franco, Tim Green, Nam Nguyen, Joe Kelley, Aroma Mahendru, Andrea Hu, Joshua Howland, Ben Vargas, Jeffrey Hui, Kshitij Bansal, Vikram Rao, Rakesh Ghiya, Emma Wang, Ke Ye, Jean Michel Sarr, Melanie Moranski Preston, Madeleine Elish, Steve Li, Aakash Kaku, Jigar Gupta, Ice Pasupat, Da-Cheng Juan, Milan Someswar, Tejvi M., Xinyun Chen, Aida Amini, Alex Fabrikant, Eric Chu, Xuanyi Dong, Amruta Muthal, Senaka Buthpitiya, Sarthak Jauhari, Nan Hua, Urvashi Khandelwal, Ayal Hitron, Jie Ren, Larissa Rinaldi, Shahar Drath, Avigail Dabush, Nan-Jiang Jiang, Harshal Godhia, Uli Sachs, Anthony Chen, Yicheng Fan, Hagai Taitelbaum, Hila Noga, Zhuyun Dai, James Wang, Chen Liang, Jenny Hamer, Chun-Sung Ferng, Chenel Elkind, Aviel Atias, Paulina Lee, VΓ­t ListΓ­k, Mathias Carlen, Jan van de Kerkhof, Marcin Pikus, Krunoslav Zaher, Paul MΓΌller, Sasha Zykova, Richard Stefanec, Vitaly Gatsko, Christoph Hirnschall, Ashwin Sethi, Xingyu Federico Xu, Chetan Ahuja, Beth Tsai, Anca Stefanoiu, Bo Feng, Keshav Dhandhania, Manish Katyal, Akshay Gupta, Atharva Parulekar, Divya Pitta, Jing Zhao, Vivaan Bhatia, Yashodha Bhavnani, Omar Alhadlaq, Xiaolin Li, Peter Danenberg, Dennis Tu, Alex Pine, Vera Filippova, Abhipso Ghosh, Ben Limonchik, Bhargava Urala, Chaitanya Krishna Lanka, Derik Clive, Yi Sun, Edward Li, Hao Wu, Kevin Hongtongsak, Ianna Li, Kalind Thakkar, Kuanysh Omarov, Kushal Majmundar, Michael Alverson, Michael Kucharski, Mohak Patel, Mudit Jain, Maksim Zabelin, Paolo Pelagatti, Rohan Kohli, Saurabh Kumar, Joseph Kim, Swetha Sankar, Vineet Shah, Lakshmi Ramachandruni, Xiangkai Zeng, Ben Bariach, Laura Weidinger, Tu Vu, Alek Andreev, Antoine He, Kevin Hui, Sheleem Kashem, Amar Subramanya, Sissie Hsiao, Demis Hassabis, Koray Kavukcuoglu, Adam Sadovsky, Quoc Le, Trevor Strohman, Yonghui Wu, Slav Petrov, Jeffrey Dean, Oriol Vinyals2023-12-19capabilitiestraining+31 articleView β†’
πŸ”— WebGemma ScopeGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGemma Scope 2Google DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”sparse-autoencodersfeatures+12 articlesView β†’
πŸ“„ Papergeneral principles for Constitutional AIarXivβ˜…β˜…β˜…β˜†β˜†SummarySandipan Kundu, Yuntao Bai, Saurav Kadavath, Amanda Askell, Andrew Callahan, Anna Chen, Anna Goldie, Avital Balwit, Azalia Mirhoseini, Brayden McLean, Catherine Olsson, Cassie Evraets, Eli Tran-Johnson, Esin Durmus, Ethan Perez, Jackson Kernion, Jamie Kerr, Kamal Ndousse, Karina Nguyen, Nelson Elhage, Newton Cheng, Nicholas Schiefer, Nova DasSarma, Oliver Rausch, Robin Larson, Shannon Yang, Shauna Kravec, Timothy Telleen-Lawton, Thomas I. Liao, Tom Henighan, Tristan Hume, Zac Hatfield-Dodds, SΓΆren Mindermann, Nicholas Joseph, Sam McCandlish, Jared Kaplan2023-10-20alignmenttruthfulness+11 articleView β†’
πŸ”— WebGenerative AI Interim Measuresβ€”Noneβ€”β€”regulationchina+11 articleView β†’
πŸ›οΈ GovernmentGenerative AI Profile (AI 600-1)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGeneva talks in May 2024β€”Noneβ€”β€”regulationchina+11 articleView β†’
πŸ”— WebGeoffrey Hintonβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGeoffrey IrvingGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”iterated-amplificationscalable-oversight+11 articleView β†’
πŸ”— WebGeorgetown CSETCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGeorgetown CSET AnalysisCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”compute-thresholdsgovernance+11 articleView β†’
πŸ”— WebGeorgetown CSET reportCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebGitHubβ€”Noneβ€”β€”software-engineeringcode-generation+32 articlesView β†’
πŸ”— WebGitHub CopilotGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebGitHub Copilot metricsβ€”Noneβ€”β€”capabilitythreshold+1β€”View β†’
πŸ”— WebGitHub Copilot Securityβ€”Noneβ€”β€”cybersecuritytimeline+11 articleView β†’
πŸ”— WebGitHub Copilot studiesβ€”Noneβ€”β€”causal-modelcorrigibility+11 articleView β†’
πŸ”— WebGitHub Developer Surveyβ€”Noneβ€”β€”human-agencyautomation+11 articleView β†’
πŸ”— WebGitHub RepositoryGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”interpretabilitycausal-scrubbing+11 articleView β†’
πŸ”— WebGiveWellβ€”Noneβ€”β€”cost-effectivenessresearch-priorities+33 articlesView β†’
πŸ”— WebGiveWell MIRI Reviewβ€”Noneβ€”β€”agent-foundationsdecision-theory+11 articleView β†’
πŸ”— WebGlassdoorβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGlassdoor Data Scientistβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGlassdoor ML Engineerβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGlassdoor Senior Data Scientistβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGlobal AI Safety Fellowshipβ€”Noneβ€”β€”safetyβ€”View β†’
πŸ“„ PaperGlobal Constitutional AIarXivβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGlobal Digital Compactβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGlobal Expansion of AI SurveillanceCarnegie Endowment

A comprehensive study reveals the widespread adoption of AI surveillance technologies worldwide, with Chinese companies playing a major role in supplying these systems to governments across different political regimes.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”authoritarianismhuman-rights+13 articlesView β†’
πŸ”— Webglobal governance researchβ€”Noneβ€”β€”governance1 articleView β†’
πŸ”— Webglobal market for AI in drug discoveryβ€”Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ”— WebGlobal Terrorism Databaseβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webglobalhealth.harvard.eduβ€”Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ”— WebGnosis Conditional Tokensβ€”Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebGoddard et al. (2012)ScienceDirectβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ”— WebGolder & Tellisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGoldman Sachsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGoldman Sachs: AI and the Global Workforce

Goldman Sachs Research predicts AI will have a limited, transitory impact on employment, with potential job displacement offset by new technological opportunities.

β€”Fullβ€”β€”economiclabor-markets+21 articleView β†’
πŸ”— WebGood 1965β€”Noneβ€”β€”power-seekingself-preservation+11 articleView β†’
πŸ”— WebGood Judgment Open

Good Judgment Open is an online forecasting platform where users can predict future events and compete to become 'Superforecasters'. The platform is operated by Good Judgment, a forecasting services firm co-founded by Philip Tetlock.

β€”Fullβ€”β€”prioritizationresource-allocation+33 articlesView β†’
πŸ”— WebGood Judgment Openβ€”Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
✏️ Blog PostGoodhart's LawAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”specification-gaminggoodharts-law+11 articleView β†’
πŸ“„ PaperGoodhart's Law empirically confirmedarXivβ˜…β˜…β˜…β˜†β˜†SummaryDavid Manheim, Scott Garrabrant2018-03-13alignmentgovernance+13 articlesView β†’
πŸ”— WebGoogleβ€”Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— WebGoogleβ€”Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebGoogle acquires DeepMindβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebGoogle collaborated on C2PA version 2.1Google AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebGoogle DeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+311 articlesView β†’
πŸ”— WebGoogle DeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-ai-interactionai-control+32 articlesView β†’
πŸ”— WebGoogle DeepMind researchersSpringerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”truthepistemology+11 articleView β†’
πŸ”— WebGoogle DeepMind: Frontier Safety Framework Version 3.0β€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebGoogle DeepMind: Introducing the Frontier Safety FrameworkGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety2 articlesView β†’
πŸ”— WebGoogle DeepMind: Strengthening our Frontier Safety FrameworkGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebGoogle DeepMind's AGI frameworkGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”agiscaling+21 articleView β†’
πŸ”— WebGoogle Fact Check Tools

I apologize, but the provided source content appears to be incomplete, fragmented, and lacks substantial information about fact-checking tools. Without a coherent document to analyze, I cannot generate a meaningful summary. For a proper analysis, I would need: - A complete source document - Clear context about the fact-checking tools - Specific details about methodology, findings, or implications If you have a more complete source document, I'd be happy to help you summarize it using the requested JSON format. Would you like to provide the full source document or clarify the content you want me to analyze?

β€”Summaryβ€”β€”β€”β€”View β†’
πŸ”— WebGoogle is shipping Gemini models faster than its AI safety reportsTechCrunch

Google is accelerating its AI model releases, including Gemini 2.5 Pro and 2.0 Flash, but has not published required safety documentation. This raises concerns about transparency and responsible AI development.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”safetyopen-source+11 articleView β†’
πŸ”— WebGoogle processes 8.5 billion searches dailyGoogle AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebGoogle SAIF 2.0Google AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”computer-usefunction-calling+11 articleView β†’
πŸ”— WebGoogle ScholarGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cognitive-emulationcoem+11 articleView β†’
πŸ”— WebGoogle ScholarGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-forecastingcompute-trends+33 articlesView β†’
πŸ”— WebGoogle ScholarGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”iterated-amplificationscalable-oversight+11 articleView β†’
πŸ”— WebGoogle ScholarGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”adversarial-robustnessml-safety+11 articleView β†’
πŸ”— WebGoogle Scholar searchGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-forecastingcompute-trends+11 articleView β†’
πŸ”— WebGoogle SynthIDGoogle DeepMind

SynthID embeds imperceptible watermarks in AI-generated content to help identify synthetic media without degrading quality. It works across images, audio, and text platforms.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”disinformationinfluence-operations+12 articlesView β†’
πŸ”— WebGoogle Transparency Reportβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebGoogle Trends

Analysis of Google search trends shows increasing public curiosity about AI's practical applications across various fields like coding, writing, and image generation.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebGoogle's 2025 Research BreakthroughsGoogle AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebGoogle's DeepMind spent an estimated $650 millionβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebGoogle's Gemini 2.5 Pro missing key safety report in violation of promisesFortune

Google launched Gemini 2.5 Pro without publishing a required safety report, contradicting previous commitments made to government and international bodies about model transparency and safety evaluations.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”safetyevaluation+11 articleView β†’
πŸ”— WebGoogle's Pathways architectureGoogle AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebGoogle's rushed Bard launchGoogle AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorgame-theory+32 articlesView β†’
πŸ”— WebGorwa et al. (2020)β€”SummaryAleksandra Urman, Stefan Katz2020-08-20human-ai-interactionai-control+11 articleView β†’
πŸ”— WebGorwa et al.: "Algorithmic Content Moderation"SAGE Journalsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ›οΈ Governmentgov.ukUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”defensesecurity+34 articlesView β†’
πŸ›οΈ GovernmentGovAICentre for the Governance of AI

A research organization focused on understanding AI's societal impacts, governance challenges, and policy implications across various domains like workforce, infrastructure, and public perception.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governanceagentic+313 articlesView β†’
πŸ›οΈ GovernmentGovAI research paper on the Baruch PlanCentre for the Governance of AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentGovAI's research on KYC schemes for compute providersCentre for the Governance of AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”compute1 articleView β†’
πŸ”— Webgovernance approaches to securing frontier AIRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governance1 articleView β†’
πŸ›οΈ GovernmentGovernance researchCentre for the Governance of AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governanceinterventions+32 articlesView β†’
πŸ›οΈ GovernmentGovernance.ai: Open Problems in Technical AI Governanceβ€”Noneβ€”β€”governance1 articleView β†’
πŸ”— WebGovernment advisory workApollo Researchβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deceptionscheming+11 articleView β†’
πŸ›οΈ Governmentgovernment AI policiesUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskeffective-altruism+38 articlesView β†’
πŸ”— WebGovernment of Canada AIDA Companion Documentβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGovernor Newsom vetoed the bill on September 29, 2024TechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGovernor Polis signs SB 24-205β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGovLab

GovLab is a research initiative focusing on transforming governance through technology, data collaboration, and citizen participation. They develop projects and resources to enhance lawmaking, responsible data use, and innovative governance approaches.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ”— WebGPAIβ€”Noneβ€”β€”governancepower-dynamics+12 articlesView β†’
πŸ”— WebGPAI Code of Practice ConsultationEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”regulationgpai+11 articleView β†’
πŸ”— WebGPQA Diamondβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGPT-4OpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llmdisinformation+22 articlesView β†’
πŸ”— WebGPT-4 scored in the top 10% on a simulated bar examSSRNβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”llm1 articleView β†’
πŸ“„ PaperGPT-4 successfully shifting political opinionsarXivβ˜…β˜…β˜…β˜†β˜†SummaryJeremy Heyl, Denis GonzΓ‘lez-Caniulef, Ilaria Caiazzo2023-11-13llmsocial-engineering+21 articleView β†’
πŸ”— WebGPT-4 technical reportβ€”Noneβ€”β€”llmscaling+21 articleView β†’
πŸ”— WebGPT-4.1 Announcement - OpenAIOpenAI

OpenAI introduces GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano models with enhanced performance across coding, instruction following, and long-context understanding. The models offer improved reliability and efficiency at lower costs.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiesllmβ€”View β†’
πŸ”— WebGPT-4.5 vs Claude 3.7 - Advanced Redteaming Analysis

VirtueAI conducted comprehensive red-teaming tests on GPT-4.5 and Claude 3.7, evaluating their performance across multiple safety and security domains. The analysis reveals distinct strengths and weaknesses in hallucination, compliance, privacy, and bias mitigation.

β€”Fullβ€”β€”capabilitiessafety+3β€”View β†’
πŸ”— WebGPT-4oOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llmgovernance+22 articlesView β†’
πŸ”— WebGPTZeroβ€”Noneβ€”β€”llm1 articleView β†’
πŸ”— WebGraham Allison's analysisβ€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebGrand View Researchβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGrand View Research - Artificial Intelligence in Military Market Reportβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentGrants OverviewUK AI Safety Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancegovernment-ai-safety+11 articleView β†’
πŸ”— WebGraph Networks for Materials Exploration (GNoME)Google DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ”— WebGraphika

Graphika offers an AI-powered platform for monitoring social media threats, detecting influence operations, and providing actionable intelligence for organizations across various sectors.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebGraphite's analysisβ€”Noneβ€”β€”truthepistemology+11 articleView β†’
πŸ›οΈ GovernmentGrassley Introduces AI Whistleblower Protection Actβ€”Noneβ€”β€”frontier-labssafety-culture+11 articleView β†’
πŸ”— WebGreat Firewall 2.0Atlantic Councilβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebGround Newsβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— Webgrowing awarenessPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebGryphon Scientificβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebGryphon Scientificβ€”Noneβ€”β€”monitoringearly-warning+11 articleView β†’
πŸ“„ PaperGSM8KarXivβ˜…β˜…β˜…β˜†β˜†SummaryKarl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, Christopher Hesse, John Schulman2021-10-27capabilitiestraining+31 articleView β†’
πŸ”— WebGSM8K Leaderboardβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebGuardianβ€”Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ“„ PaperGuess et al., Science Advances (2023)Scienceβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ›οΈ GovernmentGuidelines and standardsNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”interventionseffectiveness+33 articlesView β†’
πŸ”— WebHackerOneβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperHadfield-Menell et al. (2016)arXivβ˜…β˜…β˜…β˜†β˜†SummaryDylan Hadfield-Menell, Anca Dragan, Pieter Abbeel, Stuart Russell2016-06-09alignmentinverse-reinforcement-learning+22 articlesView β†’
πŸ“„ PaperHadfield-Menell et al. (2017)arXivβ˜…β˜…β˜…β˜†β˜†SummaryDylan Hadfield-Menell, Anca Dragan, Pieter Abbeel, Stuart Russell2016-11-24capabilitiessafety+33 articlesView β†’
πŸ“„ PaperHagendorff et al. 2024arXivβ˜…β˜…β˜…β˜†β˜†SummaryPeter S. Park, Simon Goldstein, Aidan O'Gara, Michael Chen, Dan Hendrycks2023-08-28governancedeception+31 articleView β†’
πŸ”— WebHaize Labsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebHanson (2003)β€”Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebHanson (2013): Futarchy

Futarchy is an alternative governance model where elected representatives define national welfare metrics, and market speculators propose policies expected to maximize those metrics through betting markets.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ”— Webhardware-enabled governance mechanismsRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecompute2 articlesView β†’
πŸ”— WebHarmBenchβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebHarvard Kennedy School Misinformation Review articleβ€”Noneβ€”β€”truthepistemology+11 articleView β†’
πŸ”— WebHarvard's Ash Centerβ€”Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ”— WebHeim et al. (2023)CSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factordiffusion+11 articleView β†’
πŸ“„ PaperHeim et al. 2024arXivβ˜…β˜…β˜…β˜†β˜†SummaryCaleb Rotello, Peter Graf, Matthew Reynolds, Eric B. Jones, Cody James Winkleblack, Wesley Jones2024-02-23game-theorygovernance+11 articleView β†’
πŸ“„ PaperHendrycks and Gimpel (2017)arXivβ˜…β˜…β˜…β˜†β˜†SummaryDan Hendrycks, Kevin Gimpel2016-10-07capabilitieseconomic+31 articleView β†’
πŸ“„ PaperHendrycks et al.arXivβ˜…β˜…β˜…β˜†β˜†SummaryDan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, Jacob Steinhardt2020-09-07capabilitiesevaluation+33 articlesView β†’
πŸ“„ PaperHierarchicalRLarXivβ˜…β˜…β˜…β˜†β˜†SummaryTejas D. Kulkarni, Karthik R. Narasimhan, Ardavan Saeedi, Joshua B. Tenenbaum2016-04-20governanceagentic+21 articleView β†’
πŸ›οΈ GovernmentHigh-frequency trading algorithmsβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebHiroshima AI Processβ€”Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— Webhistorical analogues researchRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebHive Moderation

I apologize, but the provided source content appears to be incomplete and consists only of a Google Tag Manager iframe snippet, which is not a substantive document. Without meaningful text describing Hive Moderation, I cannot generate a comprehensive summary. To properly analyze this source, I would need: 1. A full text document about Hive Moderation 2. Context about the topic 3. Substantive content describing its purpose, methodology, or findings If you have additional details or the full text of the document, I'd be happy to help you create the requested summary and analysis. Would you like to: - Provide the full source document - Clarify the source content - Confirm if this is the complete text

β€”Summaryβ€”β€”β€”β€”View β†’
πŸ”— WebHKSβ€”Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ“„ PaperHo et al. 2024arXivβ˜…β˜…β˜…β˜†β˜†SummaryHans Gundlach, Alex Fogelson, Jayson Lynch, Ana Trisovic, Jonathan Rosenfeld, Anmol Sandhu, Neil Thompson2025-11-26capabilitiestraining+21 articleView β†’
πŸ“„ PaperHoffmann et al. (2022)arXivβ˜…β˜…β˜…β˜†β˜†SummaryJordan Hoffmann, Sebastian Borgeaud, Arthur Mensch, Elena Buchatskaya, Trevor Cai, Eliza Rutherford, Diego de Las Casas, Lisa Anne Hendricks, Johannes Welbl, Aidan Clark, Tom Hennigan, Eric Noland, Katie Millican, George van den Driessche, Bogdan Damoc, Aurelia Guy, Simon Osindero, Karen Simonyan, Erich Elsen, Jack W. Rae, Oriol Vinyals, Laurent Sifre2022-03-29capabilitiestraining+35 articlesView β†’
πŸ”— WebHolistic AI: High Cost of Non-Compliance Under AI Law

Organizations face increasing legal and financial risks from AI non-compliance across jurisdictions. Penalties range from thousands to billions of euros for privacy, transparency, and algorithmic bias violations.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebHolocaust denial groupsβ€”Noneβ€”β€”historical-evidencearchives+11 articleView β†’
πŸ“„ PaperHolzinger et al.β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webhomomorphic encryptionβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebHow Racial Bias Infected a Major Health-Care Algorithmβ€”Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— WebHow to Improve AI Red-Teaming: Challenges and RecommendationsCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”benchmarksred-teaming+11 articleView β†’
πŸ”— WebHow We Analyzed the COMPAS Recidivism Algorithmβ€”Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— WebHPCwire: Breaking Down Global Government Spending on AIβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— Webhttps://mila.quebec/en/ai-safety/β€”Noneβ€”β€”safetydeep-learning+21 articleView β†’
πŸ“„ PaperHubinger et al. (2024)arXivβ˜…β˜…β˜…β˜†β˜†SummaryShanshan Han2024-10-09interventionseffectiveness+11 articleView β†’
πŸ”— WebHudson Instituteβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebHugging Faceβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ”— WebHugging Face Open Source AIβ€”Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebHugging Face's BLOOMβ€”Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebHuman CompatibleAmazonβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”intelligence-explosionrecursive-self-improvement+34 articlesView β†’
πŸ”— WebHuman CompatibleAmazonβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebHuman detection rates below chance in some studiesPNASβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”epistemictimeline+33 articlesView β†’
πŸ”— WebHuman Factors and Ergonomics Society

The Human Factors and Ergonomics Society (HFES) is a professional organization that advances the science of designing systems and technologies with human needs in mind. It provides networking, research, and professional development opportunities for experts in human factors and ergonomics.

β€”Fullβ€”β€”automationhuman-factors+11 articleView β†’
πŸ›οΈ GovernmentHuman Factors in Aviation

The FAA's human factors research focuses on understanding and improving human performance in aviation maintenance through scientific and applied studies. The research aims to reduce errors by identifying critical performance factors.

β€”Fullβ€”β€”capabilitiesautomation+21 articleView β†’
πŸ”— WebHuman performance in detecting deepfakes: A systematic review and meta-analysisScienceDirectβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitiesdeepfakes+33 articlesView β†’
πŸ”— WebHuman Rights Watchβ€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebHuman Rights Watch documentationβ€”Noneβ€”β€”authoritarianismstability+11 articleView β†’
πŸ”— WebHuman Rights Watch notesβ€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebHuman-Computer Interaction Journalβ€”Noneβ€”β€”computemental-health+21 articleView β†’
πŸ”— WebHumanEvalGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebHumanEval: When Machines Learned to Code - Runloop

OpenAI's HumanEval introduced a standardized benchmark with 164 Python programming problems to assess AI code generation performance. It established the pass@k metric and became the gold standard for measuring coding AI capabilities.

β€”Fullβ€”β€”capabilitiesevaluationβ€”View β†’
πŸ”— WebHypersense AI Adoption Trends

The 2024 AI landscape shows exponential growth across multiple sectors, with global AI spending projected to reach $500 billion and over 70% of organizations adopting AI technologies for at least one business function.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentIan HogarthUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancegovernment-ai-safety+12 articlesView β†’
πŸ”— WebIAPP AI Governanceβ€”Noneβ€”β€”governance1 articleView β†’
πŸ”— WebIAPP State AI Governance Trackerβ€”Noneβ€”β€”governance1 articleView β†’
πŸ”— WebIAPP: Global AI Law and Policy Tracker

The IAPP Global AI Law and Policy Tracker monitors AI governance initiatives worldwide, capturing legislative efforts, national strategies, and policy approaches across different jurisdictions.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ”— WebIAPS AI Reliability Survey

A comprehensive expert survey mapping out the most promising and urgent research directions in AI reliability and security. The study provides a data-driven ranking of potential research impacts and recommendations.

β€”Fullβ€”β€”cybersecurityβ€”View β†’
πŸ”— WebIAPS governance researchβ€”Noneβ€”β€”governanceinterventions+21 articleView β†’
πŸ›οΈ GovernmentIARPA forecastingβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebIATA reports

The International Air Transport Association (IATA) is a trade association representing airlines, providing industry reports and strategic services. They cover economic outlooks, market analyses, and airline industry developments.

β€”Fullβ€”β€”economicautomation+21 articleView β†’
πŸ”— WebIBM FHEβ€”Noneβ€”β€”game-theorygovernance+11 articleView β†’
πŸ”— WebIBM Securityβ€”Noneβ€”β€”cybersecuritytimeline+21 articleView β†’
πŸ”— WebIBM's 2025 Cost of a Data Breach Reportβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebIBM's Watson for Oncologyβ€”Noneβ€”β€”robustnessgeneralization+11 articleView β†’
πŸ“„ PaperICLR 2017arXivβ˜…β˜…β˜…β˜†β˜†SummaryTakeru Miyato, Andrew M. Dai, Ian Goodfellow2016-05-25capabilitiestraining+31 articleView β†’
πŸ“„ PaperICLR 2021arXivβ˜…β˜…β˜…β˜†β˜†SummaryDan Hendrycks, Collin Burns, Steven Basart, Andrew Critch, Jerry Li, Dawn Song, Jacob Steinhardt2020-08-05alignmentcapabilities+31 articleView β†’
πŸ”— WebICLR 2024β€”Noneβ€”β€”specification-gaminggoodharts-law+11 articleView β†’
πŸ”— WebICLR 2025β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebICML 2024Semantic Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“‹ ReportICML 2024 Fact Sheet

The 41st International Conference on Machine Learning (ICML) will be held in Vienna, Austria, with a comprehensive program of research presentations, workshops, and invited talks across machine learning disciplines.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebICML 2024 Statisticsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebIEA Energy and AI Report

The International Energy Agency's report analyzes the energy implications of AI, focusing on electricity demand, energy sources, and potential impacts on security, emissions, and innovation.

β€”Fullβ€”β€”cybersecurity1 articleView β†’
πŸ”— WebIEA Energy and AI Report

The International Energy Agency forecasts data center electricity consumption to double by 2030, with AI-driven accelerated servers being a key driver of increased energy demand.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebIEA projectionsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebIEEE 7000-2021β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebIEEE Ethics in AIβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebIEEE Interview 2024β€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebIEEE Spectrum: Content Credentials vs Deepfakesβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebIEEE Spectrum: Open-Source AI Dangersβ€”Noneβ€”β€”open-sourceβ€”View β†’
πŸ”— WebIEEE Standardsβ€”Noneβ€”β€”social-engineeringmanipulation+11 articleView β†’
πŸ”— WebIEEE Standardsβ€”Noneβ€”β€”escalationconflict+12 articlesView β†’
πŸ”— WebIEEE Standards Associationβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebIIM Ahmedabadβ€”Noneβ€”β€”human-agencyautomation+11 articleView β†’
πŸ”— WebIJOP systemβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebIlya Sutskeverβ€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebImageNet Classification with Deep CNNsβ€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebImageNet competitionβ€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebImageTwinβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebImageTwinβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebIMD AI Safety Clockβ€”Noneβ€”β€”safetyx-risk+32 articlesView β†’
πŸ”— WebIMFInternational Monetary Fundβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic-inequalitymarket-concentration+11 articleView β†’
πŸ”— WebIMF Economic OutlookInternational Monetary Fundβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic1 articleView β†’
πŸ”— WebIMF Future of GrowthInternational Monetary Fundβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebIMF: AI and Global EconomyInternational Monetary Fundβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”labor-marketsautomation+13 articlesView β†’
πŸ”— WebIMF: AI and Market VolatilityInternational Monetary Fundβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”algorithmic-tradingfinancial-stability+32 articlesView β†’
πŸ”— WebIMF: Tech's Winner-Take-All TrapInternational Monetary Fundβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebIMF's 2024 reportInternational Monetary Fundβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ”— WebIMF's October 2024 Global Financial Stability ReportInternational Monetary Fundβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ”— WebImplemented at OpenAIOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”iterated-amplificationscalable-oversight+11 articleView β†’
πŸ”— WebImplications of AI in Cybersecurityβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebImport AI Newsletterβ€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebImproving but challengedβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ“„ PaperImproving Factuality and Reasoning through Multiagent DebatearXivβ˜…β˜…β˜…β˜†β˜†SummaryYilun Du, Shuang Li, Antonio Torralba, Joshua B. Tenenbaum, Igor Mordatch2023-05-23capabilitiesllm+31 articleView β†’
πŸ“„ PaperImproving Weak-to-Strong with Scalable OversightarXivβ˜…β˜…β˜…β˜†β˜†SummaryJitao Sang, Yuhang Wang, Jing Zhang, Yanxu Zhu, Chao Kong, Junhong Ye, Shuyu Wei, Jinlin Xiao2024-02-01alignmentcapabilities+31 articleView β†’
πŸ›οΈ Governmentinaugural International AI Safety ReportUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety3 articlesView β†’
πŸ”— WebIndeedβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ Paperindependent evaluationarXivβ˜…β˜…β˜…β˜†β˜†SummaryJoeran Beel, Min-Yen Kan, Moritz Baumgart2025-02-20evaluation1 articleView β†’
πŸ”— WebInduction HeadsTransformer Circuitsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webindustry analysisMediumβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— Webindustry estimatesAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”software-engineeringcode-generation+11 articleView β†’
πŸ”— Webindustry leadersβ€”Noneβ€”β€”social-engineeringvoice-cloning+11 articleView β†’
πŸ”— WebIndustry lobbyingβ€”Noneβ€”β€”risk-factorgame-theory+11 articleView β†’
πŸ”— WebIndustry researchβ€”Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ“„ PaperIndustry standard emergingarXivβ˜…β˜…β˜…β˜†β˜†SummaryD. Estevez-Moya, E. Estevez-Rams, H. Kantz2023-10-01interpretabilityrisk-factor+21 articleView β†’
πŸ”— Webinfluenced OpenAI's o1 model seriesβ€”Noneβ€”β€”debaterecursive-reward-modeling+11 articleView β†’
πŸ“„ PaperInfoRM: Mitigating Reward Hacking in RLHFarXiv

A novel framework called InfoRM addresses reward misgeneralization in RLHF by introducing a variational information bottleneck objective to filter irrelevant reward features and detect overoptimization.

β˜…β˜…β˜…β˜†β˜†FullMiao, Yuchun, Zhang, Sen, Ding, Liang, Bao, Rong, Zhang, Lefei, Tao, Dacheng2024interpretabilitytraining+1β€”View β†’
πŸ”— WebInformation Overload ResearchGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebInfosecurity Magazine: AISI Rebrandsβ€”Noneβ€”β€”governancegovernment-ai-safety+11 articleView β†’
πŸ”— WebInfosecurity Magazine: Seoul Summit Coverageβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebInfosys researchβ€”Noneβ€”β€”β€”1 articleView β†’
✏️ Blog PostInner Alignment ProblemAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”alignmentmesa-optimization1 articleView β†’
πŸ›οΈ GovernmentInsilico Medicine's AI-designed drug candidate INS018_055β€”Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ”— WebInspect Evalsβ€”Noneβ€”β€”evaluation1 articleView β†’
πŸ”— WebInstitute for AI Policy and Strategy analysisβ€”Noneβ€”β€”governance2 articlesView β†’
πŸ”— WebInstitute for Historical Reviewβ€”Noneβ€”β€”historical-evidencearchives+11 articleView β†’
πŸ”— WebIntegraNXT ROI Analysis

An analysis of AI automation's return on investment (ROI) that explores both tangible and intangible benefits across organizational functions. The study highlights the complexity of measuring AI's comprehensive impact.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebIntel FakeCatcherβ€”Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— Webintelligence.orgMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”defensesecurity+11 articleView β†’
πŸ“„ Paperinterdisciplinary review of AI evaluationarXivβ˜…β˜…β˜…β˜†β˜†SummaryMaria Eriksson, Erasmo Purificato, Arman Noroozian, Joao Vinagre, Guillaume Chaslot, Emilia Gomez, David Fernandez-Llorca2025-02-10evaluation1 articleView β†’
πŸ”— WebInterim report publishedUnited Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ“‹ ReportIntermediate AI GovernanceMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”governance1 articleView β†’
πŸ”— Webinternal governance frameworksGoogle AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancex-risk+21 articleView β†’
πŸ›οΈ GovernmentInternational AI Safety Report (2025)UK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebInternational AI Safety Report (October 2025)β€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebInternational AI Safety Report 2025

The International AI Safety Report 2025 provides a global scientific assessment of general-purpose AI capabilities, risks, and potential management techniques. It represents a collaborative effort by 96 experts from 30 countries to establish a shared understanding of AI safety challenges.

β€”Fullβ€”β€”capabilitiessafety+37 articlesView β†’
πŸ”— WebInternational AI Safety Report 2025β€”Noneβ€”β€”safetyβ€”View β†’
πŸ“„ Paperinternational AI treatyNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebInternational Cooperation on AI GovernanceFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governance1 articleView β†’
πŸ”— WebInternational Fact-Checking Network

The IFCN supports fact-checkers worldwide through grants, training, resources, and an annual global conference. They advocate for journalistic integrity and truth-telling in media.

β€”Fullβ€”β€”trainingdisinformation+21 articleView β†’
πŸ”— WebInternational Fact-Checking Networkβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebInternational Governance of AISpringer

The article explores various governance strategies for transformative AI, analyzing potential approaches from subnational norms to international regimes. It highlights the unique challenges of governing AI due to its rapid development, dual-use potential, and complex technological landscape.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governance1 articleView β†’
πŸ›οΈ Governmentinternational network of AI Safety InstitutesUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety1 articleView β†’
πŸ›οΈ GovernmentInternational Network of AI Safety InstitutesNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”safety3 articlesView β†’
πŸ›οΈ GovernmentInternational Network of AI Safety Institutesβ€”Noneβ€”β€”safetyx-risk+23 articlesView β†’
πŸ›οΈ GovernmentInternational Network of AI Safety Institutesβ€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebInternet Archive

The source document requires JavaScript to be enabled, preventing direct content analysis.

β€”Fullβ€”β€”historical-evidencearchives+11 articleView β†’
πŸ”— WebInternet Policy Reviewβ€”Noneβ€”β€”governanceai-ethics+21 articleView β†’
πŸ”— WebINTERPOL's AI crime initiativesβ€”Noneβ€”β€”social-engineeringvoice-cloning+11 articleView β†’
πŸ”— WebinterpretabilityAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interpretability1 articleView β†’
πŸ”— WebInterviewing AI researchers on automation of AI R&DEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic1 articleView β†’
πŸ”— WebIntrospection ResearchTransformer Circuitsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deceptionself-awareness+11 articleView β†’
πŸ”— WebInvestigating LLM Jailbreaking of Popular Generative AI Web Products

A comprehensive study examining how large language models can be manipulated to bypass safety guardrails through single-turn and multi-turn jailbreak techniques. The research reveals widespread vulnerabilities across generative AI web products.

β€”Fullβ€”β€”safetycybersecurity+1β€”View β†’
πŸ”— Webiproov.comβ€”Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ”— WebIpsos

A global survey exploring public perceptions of AI, finding people are simultaneously excited and apprehensive about AI's potential impact on society and work.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebIranianMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebIREX: Learn to Discernβ€”Noneβ€”β€”β€”β€”View β†’
πŸ“„ PaperIrvin et al. (2019)arXivβ˜…β˜…β˜…β˜†β˜†SummaryJeremy Irvin, Pranav Rajpurkar, Michael Ko, Yifan Yu, Silviana Ciurea-Ilcus, Chris Chute, Henrik Marklund, Behzad Haghgoo, Robyn Ball, Katie Shpanskaya, Jayne Seekins, David A. Mong, Safwan S. Halabi, Jesse K. Sandberg, Ricky Jones, David B. Larson, Curtis P. Langlotz, Bhavik N. Patel, Matthew P. Lungren, Andrew Y. Ng2019-01-21capabilitiestraining+31 articleView β†’
πŸ”— WebIS Partners: NIST AI RMF 2025 Updates

NIST is updating its AI Risk Management Framework to provide more comprehensive guidance on AI governance, focusing on generative AI, supply chain risks, and evolving threat models.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ“„ PaperIs Power-Seeking AI an Existential Risk?arXivβ˜…β˜…β˜…β˜†β˜†SummaryJoseph Carlsmith2022-06-16alignmentx-risk+36 articlesView β†’
πŸ”— WebISACA: Understanding the EU AI Act

The EU AI Act is the first comprehensive global AI regulation, establishing requirements and risk classifications for AI systems. It aims to ensure safe, ethical, and responsible AI development and deployment.

β€”Fullβ€”β€”governancesafetyβ€”View β†’
πŸ”— WebISD Global: Online Extremismβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebISG Enterprise AI Report

The ISG Enterprise AI Report provides insights into AI adoption trends across businesses, highlighting both progress and obstacles in implementing AI solutions. The research covers 1,200 AI use cases and examines enterprise AI strategies and performance.

β€”Fullβ€”β€”capabilities1 articleView β†’
πŸ”— WebISO Standardsβ€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebISO/IEC 23894:2023β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebISO/IEC 42001β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebISO/IEC 42001 certificationβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebISO/IEC 42001:2023β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebIsomorphic Labsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebIsraeli Iron Dome autonomous interceptsβ€”Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— Webissued an internal "code red" memoTechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
✏️ Blog PostIterated AmplificationAlignment Forumβ˜…β˜…β˜…β˜†β˜†MetadataAjeya Cotra2018-11-30β€”1 articleView β†’
πŸ“„ PaperIterated Distillation and AmplificationarXivβ˜…β˜…β˜…β˜†β˜†SummaryPaul Christiano, Buck Shlegeris, Dario Amodei2018-10-19alignmentcapabilities+32 articlesView β†’
πŸ”— WebITIF Analysis

An analysis shows AI generated approximately 119,900 jobs in 2024 while causing only 12,700 job losses. The technology is reshaping workforce dynamics rather than destroying employment.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebITIF: Export Controls and American AI Leadership

The Biden and Trump administrations' restrictive export policies for AI chips are harming U.S. technology firms by cutting them off from global markets and inadvertently stimulating Chinese domestic innovation.

β€”Fullβ€”β€”computeβ€”View β†’
πŸ”— WebIttefaq et al. (2024)ScienceDirect

A comprehensive analysis of AI news coverage in 12 newspapers from 2010-2023 using topic modeling and sentiment analysis. The study reveals differences in AI framing between Global North and South media outlets.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebITU Annual AI Governance Report 2025β€”Noneβ€”β€”governance1 articleView β†’
πŸ”— WebJack Clark, Anthropicβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebJailbreakBench: LLM robustness benchmark

JailbreakBench introduces a centralized benchmark for assessing LLM robustness against jailbreak attacks, including a repository of artifacts, evaluation framework, and leaderboards.

β€”Fullβ€”β€”capabilitiesevaluation+11 articleView β†’
πŸ”— Webjailbreaking research from CMUβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ“„ PaperjailbreaksarXivβ˜…β˜…β˜…β˜†β˜†SummaryAndy Zou, Zifan Wang, Nicholas Carlini, Milad Nasr, J. Zico Kolter, Matt Fredrikson2023-07-27alignmenteconomic+23 articlesView β†’
πŸ”— WebJAMA Internal Medicineβ€”Noneβ€”β€”networksrisk-interactions+11 articleView β†’
πŸ”— WebJan Leike arguesβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebJan Leike resigns, posts "safety culture has taken a backseat to shiny products"β€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebJapan AISIβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebJason Wei of Google Brainβ€”Noneβ€”β€”scalingcapability-evaluation+11 articleView β†’
πŸ”— WebJCI: AlphaFold developers share 2023 Lasker Awardβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebJMIR Mental Health: AI in Mental Healthβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebJoe Carlsmith's comprehensive analysis of schemingOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deceptionprobability+21 articleView β†’
πŸ”— WebJohns Hopkins Center for Health Securityβ€”Noneβ€”β€”cybersecurityprobability+21 articleView β†’
πŸ”— Webjoined AnthropicTechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”research-agendasalignment+11 articleView β†’
πŸ”— WebJoseph Nye arguesβ€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebJournal of Deliberative Democracyβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebJournal of Democracyβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebJournal of Democracyβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebJournal of Finance (2024)β€”Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ”— WebJPMorgan Chase's risk assessmentβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ“„ PaperJuly 2024 in Virginia's "Data Center Alley"arXivβ˜…β˜…β˜…β˜†β˜†SummaryXin Chen, Xiaoyang Wang, Ana Colacelli, Matt Lee, Le Xie2025-09-08trainingalgorithmic-trading+21 articleView β†’
πŸ”— WebJuma GPT-4 cost breakdown

The article explores the costs of training large language models like GPT-3 and GPT-4, highlighting the substantial financial and environmental implications of AI model development.

β€”Fullβ€”β€”trainingllm1 articleView β†’
πŸ”— WebJune 30, 2026β€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebKalshi

I apologize, but the provided content does not appear to be a substantive source document. It seems to be a fragment of a webpage with some tracking code and partial menu items, but lacks any meaningful text about Kalshi or a coherent source to analyze. Without a clear, complete text describing Kalshi, its purpose, research, or contributions, I cannot responsibly complete the requested JSON summary. If you have a specific document, research paper, or detailed description about Kalshi that you would like me to analyze, please provide the full text, and I will be happy to help you create the structured summary. Would you like to share the complete source document or provide more context about the source?

β€”Summaryβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ“„ PaperKaplan & Garrick (1981)β€”Noneβ€”β€”prioritizationresource-allocation+11 articleView β†’
πŸ“„ PaperKaplan et al. (2020)arXivβ˜…β˜…β˜…β˜†β˜†SummaryJared Kaplan, Sam McCandlish, Tom Henighan, Tom B. Brown, Benjamin Chess, Rewon Child, Scott Gray, Alec Radford, Jeffrey Wu, Dario Amodei2020-01-23capabilitiestraining+36 articlesView β†’
πŸ”— WebKargu-2 autonomous engagement in LibyaUnited Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— WebKargu-2 loitering munitionβ€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ“„ PaperKasirzadeh (2024)arXivβ˜…β˜…β˜…β˜†β˜†SummaryAtoosa Kasirzadeh2024-01-15governancesafety+31 articleView β†’
πŸ“„ PaperKenton et al. (2021)arXivβ˜…β˜…β˜…β˜†β˜†SummaryStephanie Lin, Jacob Hilton, Owain Evans2021-09-08capabilitiestraining+33 articlesView β†’
πŸ“„ PaperKevin Esvelt warningsNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebKey Outcomes of the AI Seoul Summitβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebKhan Academy AI Tutor Resultsβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ”— WebKilobaserβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebKnight First Amendment Institute: Epistemic Infrastructure

The Knight First Amendment Institute focuses on legal and constitutional challenges in digital communication, with emerging research on AI's implications for democratic resilience.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebKnight Foundationβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebKnight Foundationβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebKnight Foundationβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebKnowBe4β€”Noneβ€”β€”social-engineeringvoice-cloning+11 articleView β†’
πŸ”— WebKonceptual AI Analysisβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebKorinek & Vipraβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebKosinski 2023β€”Noneβ€”β€”scalingcapability-evaluation+11 articleView β†’
πŸ”— WebKPMG Global AI Trust Study

A comprehensive survey of 48,000 people across 47 countries explores public attitudes towards AI, highlighting rising adoption and critical trust challenges.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebKPMG Venture Pulse

KPMG's Venture Pulse report highlights a global VC investment increase to $368.3 billion in 2024, with AI sector emerging as a major investment driver despite reduced deal volumes.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebKrakovna et al.β€”Noneβ€”β€”corrigibilityshutdown-problem+11 articleView β†’
πŸ“„ PaperKrakovna et al. (2020)arXivβ˜…β˜…β˜…β˜†β˜†SummaryYuan Fang, Jennifer Cano2020-02-07probabilitygeneralization+11 articleView β†’
πŸ”— WebLancet Microbe publicationβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— Weblandmark MIT studyβ€”Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ”— WebLangosco et al. (2022)β€”Noneβ€”β€”inner-alignmentdistribution-shift+12 articlesView β†’
πŸ“„ PaperLangosco et al. (2022)arXivβ˜…β˜…β˜…β˜†β˜†SummaryRohin Shah, Vikrant Varma, Ramana Kumar, Mary Phuong, Victoria Krakovna, Jonathan Uesato, Zac Kenton2022-10-04alignmentcapabilities+33 articlesView β†’
πŸ“„ PaperLangosco et al. (2022)arXivβ˜…β˜…β˜…β˜†β˜†SummaryLauro Langosco, Jack Koch, Lee Sharkey, Jacob Pfau, Laurent Orseau, David Krueger2021-05-28capabilitiesmesa-optimization+34 articlesView β†’
πŸ”— Weblarge language modelsAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llmframework+32 articlesView β†’
πŸ“„ PaperLarge language models trained on internet dataarXivβ˜…β˜…β˜…β˜†β˜†SummaryAlec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, Ilya Sutskever2021-02-26capabilitiestraining+31 articleView β†’
πŸ”— Weblargest single advantageRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ Governmentlater expandedβ€”Noneβ€”β€”self-regulationindustry-commitments+11 articleView β†’
πŸ”— Weblaw establishes a practical frameworkβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebLawfare analysis notesβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebLawfare Daily Podcast: State Senator Scott Wiener on SB 1047β€”Noneβ€”β€”regulationstate-policy+11 articleView β†’
πŸ”— WebLawfare Podcastβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebLawfare: Selling Spirals and AI Flash Crash

Gary Gensler warns that AI-driven algorithmic trading could trigger financial market crashes through synchronized, high-speed trading behaviors. The article explores potential regulatory and technical solutions to mitigate these risks.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ”— WebLawyer sanctioned for fake citationsThe New York Timesβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ”— WebLeading AI researchersAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webleaked surveillance documentsβ€”Noneβ€”β€”authoritarianismstability+11 articleView β†’
πŸ“„ PaperLearning representations by back-propagating errorsNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebLegal analysisβ€”Noneβ€”β€”compute-thresholdsgovernance+11 articleView β†’
πŸ”— Weblegal analysis by Stanford Lawβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— Weblegal scholars at Georgetown Lawβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ›οΈ GovernmentLennart Heim's researchCentre for the Governance of AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebLeopold Aschenbrenner (2024)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebLeopold Aschenbrenner's "Situational Awareness"β€”Noneβ€”β€”capability-generalizationalignment-stability+11 articleView β†’
✏️ Blog PostLessWrongLessWrongβ˜…β˜…β˜…β˜†β˜†Metadatapaulfchristiano, Mark Xu, Ajeya Cotra2021-12-14eliciting-latent-knowledgeelk+11 articleView β†’
✏️ Blog PostLessWrongLessWrongβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agent-foundationsdecision-theory+12 articlesView β†’
✏️ Blog PostLessWrong (2024). "Instrumental Convergence Wiki"LessWrongβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”power-seekingself-preservation+11 articleView β†’
✏️ Blog PostLessWrong GPU estimatesLessWrong

A detailed breakdown of expected GPU and compute availability across major tech companies like Microsoft, Meta, Google, Amazon, and XAI. Estimates are based on publicly available data and Nvidia revenue information.

β˜…β˜…β˜…β˜†β˜†FullCharlesD2024-11-28compute1 articleView β†’
✏️ Blog PostLessWrong postLessWrongβ˜…β˜…β˜…β˜†β˜†MetadataNickGabs2023-09-19β€”1 articleView β†’
✏️ Blog PostLessWrong PostsLessWrongβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”cognitive-emulationcoem+11 articleView β†’
✏️ Blog PostLessWrong SequencesLessWrongβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agent-foundationsdecision-theory+11 articleView β†’
✏️ Blog PostLessWrong surveysLessWrongβ˜…β˜…β˜…β˜†β˜†MetadataAorou2022-09-24β€”1 articleView β†’
✏️ Blog PostLessWrong: "Disentangling Corrigibility: 2015-2021"LessWrongβ˜…β˜…β˜…β˜†β˜†MetadataKoen.Holtman2021-02-16shutdown-problemai-control+11 articleView β†’
πŸ”— Webletter to Sam Altmanβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebLevels.fyi

Levels.fyi is a web platform that allows employees to anonymously share salary, compensation, and workplace insights. It provides transparent information about job roles and pay across different companies.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebLevels.fyi

Levels.fyi is a crowd-sourced salary and compensation platform that allows tech workers to share anonymous salary and job information. It provides insights into compensation trends and job market details.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebLex Fridman #266: Nicole Perlrothβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebLex Fridman #368: Eliezer Yudkowskyβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebLex Fridman #420: Annie Jacobsenβ€”Noneβ€”β€”game-theorycoordination+11 articleView β†’
πŸ”— Webliar's dividendβ€”Noneβ€”β€”truthepistemology+12 articlesView β†’
πŸ›οΈ GovernmentLibrary of Congress analysisβ€”Noneβ€”β€”regulationchina+11 articleView β†’
πŸ›οΈ GovernmentLibrary of Congress: CoE AI Treaty

A framework treaty opened for signature in September 2024, establishing broad legal commitments for responsible AI development across 46 member states and 11 non-member countries.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebLieber Instituteβ€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebLiebowitz and Margolis (1990)β€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebLimited progressUnited Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebLittler Mendelson analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebLlama 3Meta AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebLlama Guard 3Meta AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebLLaMA leakβ€”Noneβ€”β€”risk-factordiffusion+32 articlesView β†’
πŸ”— WebLlamaFirewallMeta AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebLLM Latency Benchmark by Use Cases

A detailed performance analysis of large language models (GPT-5.2, Mistral Large, Claude, Grok, DeepSeek) measuring first token and per-token latency across Q&A, summarization, translation, business analysis, and coding tasks.

β€”Fullβ€”β€”capabilitiesevaluation+1β€”View β†’
πŸ”— WebLLM Math Benchmark 2025 Results

The 2025 LLM math benchmarks reveal significant progress in mathematical reasoning capabilities across models like Gemini, Claude, and ChatGPT. Innovations in training and tool integration are driving substantial improvements in math problem-solving accuracy.

β€”Fullβ€”β€”capabilitiestraining+2β€”View β†’
πŸ”— WebLLM-Honesty-Survey (2025-TMLR)GitHub

A systematic review of honesty in Large Language Models, analyzing their ability to recognize known/unknown information and express knowledge faithfully. The survey provides a structured framework for evaluating and improving LLM trustworthiness.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”evaluationllmβ€”View β†’
πŸ“„ PaperLLM-Modulo frameworkarXivβ˜…β˜…β˜…β˜†β˜†SummarySubbarao Kambhampati, Karthik Valmeekam, Lin Guan, Mudit Verma, Kaya Stechly, Siddhant Bhambri, Lucas Saldyt, Anil Murthy2024-02-02llmdecision-theory+21 articleView β†’
πŸ“„ PaperLLMs Learn by IntrospectionarXivβ˜…β˜…β˜…β˜†β˜†SummaryFelix J Binder, James Chua, Tomek Korbak, Henry Sleight, John Hughes, Robert Long, Ethan Perez, Miles Turpin, Owain Evans2024-10-17interpretabilitycapabilities+31 articleView β†’
πŸ”— Weblogical inductionMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperLogical InductorsarXivβ˜…β˜…β˜…β˜†β˜†SummaryScott Garrabrant, Tsvi Benson-Tilsen, Andrew Critch, Nate Soares, Jessica Taylor2016-09-12evaluationcompute+31 articleView β†’
πŸ”— WebLong-Term Future Fundβ€”Noneβ€”β€”cost-effectivenessresearch-priorities+12 articlesView β†’
πŸ”— WebLongitudinal studies by UC San Diegoβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ“„ PaperLow-Rank Adaptation (LoRA)arXivβ˜…β˜…β˜…β˜†β˜†SummaryEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, Weizhu Chen2021-06-17trainingcompute+31 articleView β†’
πŸ”— WebLuthor AI: Avoiding AI-Washing - SEC Fines

The SEC is cracking down on misleading AI claims in financial marketing, targeting firms that overstate their artificial intelligence capabilities. Companies must now provide specific, substantiated documentation of their AI technologies.

β€”Fullβ€”β€”capabilitiesβ€”View β†’
πŸ“„ PaperMA-RLHFarXivβ˜…β˜…β˜…β˜†β˜†SummaryYekun Chai, Haoran Sun, Huang Fang, Shuohuan Wang, Yu Sun, Hua Wu2024-10-03governancecapabilities+31 articleView β†’
πŸ”— WebMacAskill (2022): What We Owe the Futureβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperMACHIAVELLI datasetarXivβ˜…β˜…β˜…β˜†β˜†SummaryAlexander Pan, Jun Shern Chan, Andy Zou, Nathaniel Li, Steven Basart, Thomas Woodside, Jonathan Ng, Hanlin Zhang, Scott Emmons, Dan Hendrycks2023-04-06capabilitiessafety+32 articlesView β†’
πŸ“„ PaperMACPO (Multi-Agent Constrained Policy Optimization)arXivβ˜…β˜…β˜…β˜†β˜†SummaryAnkita Kushwaha, Kiran Ravish, Preeti Lamba, Pawan Kumar2025-05-22governancesafety1 articleView β†’
πŸ”— WebMacroPolo Global AI Talent Tracker 2.0

The report tracks global AI talent distribution using NeurIPS conference paper data, examining researcher origins, destinations, and mobility trends across key countries.

β€”Fullβ€”β€”β€”1 articleView β†’
✏️ Blog PostMahaztra 2024LessWrongβ˜…β˜…β˜…β˜†β˜†Metadatafarrelmahaztra2025-02-14evaluationsdeception+11 articleView β†’
πŸ”— WebManaging AI RisksRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”causal-modelcorrigibility+32 articlesView β†’
πŸ›οΈ Governmentmandatory deepfake watermarkingWhite Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”social-engineeringvoice-cloning+11 articleView β†’
πŸ”— WebManhattan Projectβ€”Noneβ€”β€”talentfield-building+11 articleView β†’
πŸ”— WebManifold AI markets

The provided text appears to be a fragmentary list of AI-related topic tags without substantive content.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebManifold Markets

No substantive information available to summarize.

β€”Summaryβ€”β€”β€”β€”View β†’
πŸ”— WebMantic AI

Mantic is an AI startup that aims to create prediction models capable of forecasting global events with higher accuracy than human experts. The company has achieved top rankings in forecasting tournaments and seeks to improve decision-making across various sectors.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ“„ PaperMany-Shot JailbreakingarXivβ˜…β˜…β˜…β˜†β˜†SummaryMaksym Andriushchenko, Francesco Croce, Nicolas Flammarion2024-04-02alignmentsafety+31 articleView β†’
πŸ”— WebMapping the Mind of a Large Language ModelAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llminterpretability+21 articleView β†’
πŸ”— WebMarc Andreessenβ€”Noneβ€”β€”risk-factorcompetition+11 articleView β†’
✏️ Blog PostMarc Andreessenβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebMarch and September 2025β€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebMarket concentrationβ€”Noneβ€”β€”risk-interactionscompounding-effects+11 articleView β†’
πŸ”— WebMarket concentration dataβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebMarket Scoring Rulesβ€”Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebMarketing AI Institute

A McKinsey report forecasts massive economic potential for AI software and services, projecting trillion-dollar impacts across multiple industries by 2040. The analysis suggests AI could fundamentally reshape economic productivity and growth.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ“„ PaperMarkowitz (1952)β€”Noneβ€”β€”prioritizationresource-allocation+11 articleView β†’
πŸ”— WebMartin Beraja, David Yang, and Noam Yuchtmanβ€”Noneβ€”β€”authoritarianismstability+11 articleView β†’
πŸ”— WebMartin Ford on Rise of the Robots

I apologize, but the provided source content appears to be incomplete and seems like a generic YouTube/Google footer, not an actual source document about Martin Ford's work on the Rise of the Robots. Without the substantive content of the source, I cannot generate a meaningful summary. To properly analyze Martin Ford's work, I would need the actual text or key excerpts from his book or talk discussing automation, AI's economic impacts, and potential societal disruptions. If you have the full text or a more complete excerpt, I'd be happy to help you summarize it using the requested JSON format. Alternatively, I can provide a summary based on my existing knowledge about Martin Ford's book "Rise of the Robots: Technology and the Threat of a Jobless Future" if that would be helpful. Would you like me to: 1. Wait for a complete source text 2. Summarize from my existing knowledge about the book 3. Clarify the source document you intended to share

β€”Summaryβ€”β€”economiclabor-markets+21 articleView β†’
πŸ”— WebMASK BenchmarkGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”capabilitiesevaluation1 articleView β†’
πŸ”— Webmatched OpenAI's o1 performanceβ€”Noneβ€”β€”capabilities1 articleView β†’
πŸ›οΈ GovernmentMaterials Projectβ€”Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ“„ PaperMATHarXivβ˜…β˜…β˜…β˜†β˜†SummaryDan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, Jacob Steinhardt2021-03-05capabilitieseconomic+33 articlesView β†’
πŸ”— WebMATS Research Program

MATS is an intensive training program that helps researchers transition into AI safety, providing mentorship, funding, and community support. Since 2021, over 446 researchers have participated, producing 150+ research papers and joining leading AI organizations.

β€”Fullβ€”β€”safetytraining+36 articlesView β†’
✏️ Blog PostMATS Spring 2024 Extension RetrospectiveLessWrongβ˜…β˜…β˜…β˜†β˜†MetadataHenningB, Matthew Wearden, Cameron Holmes, Ryan Kidd2025-02-12field-buildingtraining-programs+11 articleView β†’
πŸ”— WebMatz et al. (2017)PNASβ˜…β˜…β˜…β˜…β˜…MetadataMatz, S. C., Kosinski, M., Nave, G., Stillwell, D. J.2017ai-ethicspersuasion+11 articleView β†’
πŸ›οΈ GovernmentMay 2024 Seoul AI Safety SummitUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetygovernance+21 articleView β†’
πŸ”— WebMay 2025 update (R1-0528)β€”Noneβ€”β€”decision-theoryepistemics+11 articleView β†’
πŸ›οΈ GovernmentMay 6, 2010 Flash Crashβ€”Noneβ€”β€”escalationconflict+32 articlesView β†’
πŸ”— WebMcInnes Cooper Key Lessonsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMcKinseyMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”software-engineeringcode-generation+32 articlesView β†’
πŸ”— WebMcKinsey 2016McKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebMcKinsey AI IndexMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebMcKinsey AI ReportMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebMcKinsey Economic Potential of GenAIMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”economic1 articleView β†’
πŸ”— WebMcKinsey EstimatesMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”economic-inequalitymarket-concentration+12 articlesView β†’
πŸ”— WebMcKinsey finds 57%McKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”labor-marketsautomation+12 articlesView β†’
πŸ”— WebMcKinsey Future of WorkMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMcKinsey Global InstituteMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”monitoringearly-warning+11 articleView β†’
πŸ”— WebMcKinsey Global InstituteMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ”— WebMcKinsey Global InstituteMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”labor-marketsautomation+11 articleView β†’
πŸ”— WebMcKinsey ReportsMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMcKinsey SaaS AI EraMcKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMcKinsey State of AI

The McKinsey report examines the transformative potential of AI technologies, highlighting their growing adoption and impact on business processes and workforce dynamics.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebMcKinsey State of AI 2024McKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebMcKinsey State of AI 2025McKinsey & Companyβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”networksrisk-interactions+13 articlesView β†’
πŸ“„ PaperMe, Myself, and AI: SAD BenchmarkarXivβ˜…β˜…β˜…β˜†β˜†SummaryRudolf Laine, Bilal Chughtai, Jan Betley, Kaivalya Hariharan, Jeremy Scheurer, Mikita Balesni, Marius Hobbhahn, Alexander Meinke, Owain Evans2024-07-05capabilitiesevaluation1 articleView β†’
πŸ”— WebMeasuring AI Ability to Complete Long Tasks - METRMETR

Research by METR demonstrates that AI models' ability to complete tasks is exponentially increasing, with task completion time doubling approximately every 7 months. This metric provides insights into AI's real-world capability progression.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiesβ€”View β†’
πŸ“„ PaperMeasuring AI Long Tasks - arXivarXivβ˜…β˜…β˜…β˜†β˜†SummaryThomas Kwa, Ben West, Joel Becker, Amy Deng, Katharyn Garcia, Max Hasin, Sami Jawhar, Megan Kinniment, Nate Rush, Sydney Von Arx, Ryan Bloom, Thomas Broadley, Haoxing Du, Brian Goodrich, Nikola Jurkovic, Luke Harold Miles, Seraphina Nix, Tao Lin, Neev Parikh, David Rein, Lucas Jun Koba Sato, Hjalmar Wijk, Daniel M. Ziegler, Elizabeth Barnes, Lawrence Chan2025-03-18capabilitiessafety+3β€”View β†’
πŸ“„ PaperMeasuring and Improving Constitutional AdherencearXivβ˜…β˜…β˜…β˜†β˜†SummaryNorman Di Palo, Edward Johns2023-12-19alignment1 articleView β†’
πŸ“„ PaperMeasuring Progress on Scalable Oversight for Large Language ModelsarXivβ˜…β˜…β˜…β˜†β˜†SummarySamuel R. Bowman, Jeeyoon Hyun, Ethan Perez, Edwin Chen, Craig Pettit, Scott Heiner, KamilΔ— LukoΕ‘iΕ«tΔ—, Amanda Askell, Andy Jones, Anna Chen, Anna Goldie, Azalia Mirhoseini, Cameron McKinnon, Christopher Olah, Daniela Amodei, Dario Amodei, Dawn Drain, Dustin Li, Eli Tran-Johnson, Jackson Kernion, Jamie Kerr, Jared Mueller, Jeffrey Ladish, Joshua Landau, Kamal Ndousse, Liane Lovitt, Nelson Elhage, Nicholas Schiefer, Nicholas Joseph, NoemΓ­ Mercado, Nova DasSarma, Robin Larson, Sam McCandlish, Sandipan Kundu, Scott Johnston, Shauna Kravec, Sheer El Showk, Stanislav Fort, Timothy Telleen-Lawton, Tom Brown, Tom Henighan, Tristan Hume, Yuntao Bai, Zac Hatfield-Dodds, Ben Mann, Jared Kaplan2022-11-04capabilitiessafety+31 articleView β†’
πŸ”— WebMeasuring Real-World Task Performance - OpenAIOpenAI

GDPval is a new evaluation framework assessing AI models' capabilities on economically valuable tasks across 44 occupations. It provides a realistic measure of how AI can support professional work across different industries.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiesevaluation+1β€”View β†’
πŸ“„ PaperMechanistic interpretabilityarXivβ˜…β˜…β˜…β˜†β˜†SummaryLeonard Bereska, Efstratios Gavves2024-04-22interpretability1 articleView β†’
πŸ”— WebMechanistic InterpretabilityAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interpretability1 articleView β†’
πŸ”— WebMechanistic InterpretabilityTransformer Circuitsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interpretabilitymesa-optimization+36 articlesView β†’
πŸ”— WebMechanistic Interpretability for AI Safety β€” A Reviewβ€”Noneβ€”β€”interpretabilitysafety+35 articlesView β†’
πŸ”— WebMechanistic interpretability workβ€”Noneβ€”β€”interpretabilitymesa-optimization+21 articleView β†’
πŸ”— WebMechanistic Interpretability Workshop at NeurIPS 2025β€”Noneβ€”β€”interpretabilitysparse-autoencoders+21 articleView β†’
πŸ”— WebMedia Literacy Indexβ€”Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ“„ PaperMedQAarXivβ˜…β˜…β˜…β˜†β˜†SummaryDi Jin, Eileen Pan, Nassim Oufattole, Wei-Hung Weng, Hanyi Fang, Peter Szolovits2020-09-28capabilitythreshold+11 articleView β†’
πŸ”— WebMegviiβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ“„ PaperMeinke et al. 2025arXivβ˜…β˜…β˜…β˜†β˜†SummarySebastian Farquhar, Vikrant Varma, David Lindner, David Elson, Caleb Biddulph, Ian Goodfellow, Rohin Shah2025-01-22alignmentsafety+31 articleView β†’
πŸ“„ PaperMelanie Mitchell: Why AI Is Harder Than We ThinkarXivβ˜…β˜…β˜…β˜†β˜†SummaryMelanie Mitchell2021-04-26β€”1 articleView β†’
πŸ“„ PaperMemGPTarXivβ˜…β˜…β˜…β˜†β˜†SummaryCharles Packer, Sarah Wooders, Kevin Lin, Vivian Fang, Shishir G. Patil, Ion Stoica, Joseph E. Gonzalez2023-10-12capabilitiesevaluation+31 articleView β†’
πŸ“„ PaperMeng et al., 2023arXivβ˜…β˜…β˜…β˜†β˜†SummaryKevin Meng, David Bau, Alex Andonian, Yonatan Belinkov2022-02-10evaluationllm+31 articleView β†’
✏️ Blog Postmesa-optimizationAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”mesa-optimizationcapability-generalization+22 articlesView β†’
πŸ”— WebMetaMeta AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorcompetition+11 articleView β†’
πŸ”— WebMetaβ€”Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— WebMeta AI ResearchMeta AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ”— WebMeta AI's Self-Rewarding Language ModelsFuture of Life Instituteβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”llm1 articleView β†’
πŸ”— WebMeta Llama 2 open-sourceMeta AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”open-sourcerisk-factor+36 articlesView β†’
πŸ”— WebMeta Oversight Boardβ€”Noneβ€”β€”disinformationinfluence-operations+12 articlesView β†’
πŸ”— WebMeta Oversight Board on AI Content Moderationβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebMeta Stable SignatureMeta AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebMeta Threat Reportsβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebMeta, OpenAI, and House Speaker Nancy Pelosi opposed the billCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMeta: Open Source AI Path Forwardβ€”Noneβ€”β€”open-sourceβ€”View β†’
πŸ”— WebMeta's $15+ billion annual AI infrastructure spendingReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebMeta's 2024 election integrity effortsβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebMeta's content moderation systemMeta AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ”— WebMeta's internal researchβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebMeta's LLaMA releasesβ€”Noneβ€”β€”open-sourcegovernance+21 articleView β†’
πŸ”— WebMeta's Make-A-Videoβ€”Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebMeta's platforms generate 4 billion social interactions dailyβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebMetaculusMetaculus

Metaculus is an online forecasting platform that allows users to predict future events and trends across areas like AI, biosecurity, and climate change. It provides probabilistic forecasts on a wide range of complex global questions.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”biosecurityprioritization+37 articlesView β†’
πŸ”— WebMetaculusMetaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebMetaculusMetaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMetaculus (Dec 2024)Metaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMetaculus AGI forecastsMetaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agi1 articleView β†’
πŸ”— WebMetaculus AGI QuestionMetaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agiβ€”View β†’
πŸ”— WebMetaculus AGI questionsMetaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agiforecasting+21 articleView β†’
πŸ”— WebMetaculus AI ForecastingMetaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMetaculus AI forecastsMetaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”timelinecapability+11 articleView β†’
πŸ”— WebMetaculus AI questionsMetaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebMetaculus prediction marketsMetaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebMetaculus resolution councilMetaculusβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— Webmetagenomic surveillanceβ€”Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ”— WebMetasploit AI modulesβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebMETRMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interventionseffectiveness+11 articleView β†’
πŸ“– ReferenceMETR - WikipediaWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”evaluationsdangerous-capabilities+11 articleView β†’
πŸ”— WebMETR o3 Evaluationβ€”Noneβ€”β€”evaluation1 articleView β†’
πŸ”— WebMETR PublicationsMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMETR: Common Elements of Frontier AI Safety PoliciesMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetydeception+23 articlesView β†’
πŸ”— WebMETR: Responsible Scaling PoliciesMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilities2 articlesView β†’
πŸ”— Webmetr.orgMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”software-engineeringcode-generation+322 articlesView β†’
πŸ”— Webmetr.orgMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationsdangerous-capabilities+11 articleView β†’
πŸ”— Webmetr.orgMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationsdangerous-capabilities+11 articleView β†’
πŸ”— WebMETR's analysisMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebMETR's analysis of 12 companiesMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationsdangerous-capabilities+15 articlesView β†’
πŸ”— WebMETR's Autonomy Evaluation Resources (March 2024)METRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationinterpretability+21 articleView β†’
πŸ”— WebMETR's June 2025 evaluationMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluation1 articleView β†’
πŸ”— WebMETR's researchMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebMicrosoftMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMicrosoftMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— WebMicrosoft and investors press for reinstatementβ€”Noneβ€”β€”frontier-labssafety-culture+11 articleView β†’
πŸ”— WebMicrosoft Azureβ€”Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— WebMicrosoft extends OpenAI investment to $10B+Microsoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebMicrosoft researchMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebMicrosoft ResearchMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— WebMicrosoft SEALGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”game-theorygovernance+11 articleView β†’
πŸ”— WebMicrosoft Security Copilot agentsMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebMicrosoft Security IntelligenceMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecuritysocial-engineering+21 articleView β†’
πŸ”— WebMicrosoft Threat Analysis CenterMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebMicrosoft VALL-EMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— WebMicrosoft Video Authenticator

Microsoft introduces Video Authenticator, a technology that analyzes media to detect artificial manipulation, alongside partnerships and media literacy efforts to combat disinformation.

β€”Fullβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebMicrosoft Video AuthenticatorMicrosoftβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— WebMicrosoft's $13+ billion investment in OpenAIβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebMicrosoft's Azure AIβ€”Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebMicrosoft's initial $1B OpenAI investmentβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebMicrosoft's Threat Analysis Centerβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebMidjourney v6β€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebMike Knoop's analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMilaβ€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebMILAβ€”Noneβ€”β€”governancecoordination+11 articleView β†’
πŸ›οΈ GovernmentMinnesota DEED Automation Study

SHRM research analyzed job automation risk using worker-reported data, finding that 19.2 million U.S. jobs are at high or very high risk of automation. Risk varies significantly by occupation and industry.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebMintz: AI Provisions in Biden's FY 2025 Budget

The budget proposal includes significant funding for AI initiatives across multiple government departments, focusing on research, safety, and talent acquisition.

β€”Fullβ€”β€”safetyβ€”View β†’
πŸ”— WebMIRIMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”capability-generalizationalignment-stability+12 articlesView β†’
πŸ”— WebMIRI BlogMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agent-foundationsdecision-theory+11 articleView β†’
πŸ”— WebMIRI PapersMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agent-foundationsdecision-theory+32 articlesView β†’
πŸ”— WebMIRI positionMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”prioritizationworldview+11 articleView β†’
πŸ”— WebMIRI research updatesMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”interventionseffectiveness+11 articleView β†’
πŸ”— WebMIRI Technical ReportsMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”networksrisk-interactions+11 articleView β†’
πŸ”— Webmiri.orgMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”software-engineeringcode-generation+317 articlesView β†’
πŸ”— WebMIRI's 2024 assessmentMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”interpretabilityscalable-oversight+13 articlesView β†’
πŸ”— WebMIRI's 2024 End-of-Year UpdateMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”research-agendasalignment+32 articlesView β†’
πŸ”— WebMIRI's recursive self-improvement analysisMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”causal-modelcorrigibility+11 articleView β†’
πŸ”— WebMIRI's theoretical work on deceptionMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”deception1 articleView β†’
✏️ Blog PostMIRI/Open Philanthropy exchange on decision theoryAlignment Forumβ˜…β˜…β˜…β˜†β˜†MetadataRob Bensinger2021-08-25β€”1 articleView β†’
πŸ”— WebMistralβ€”Noneβ€”β€”risk-factordiffusion+11 articleView β†’
πŸ”— WebMIT AI Incident Trackerβ€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebMIT CSAILβ€”Noneβ€”β€”social-engineeringmanipulation+11 articleView β†’
πŸ”— WebMIT CSAILβ€”Noneβ€”β€”timelineautomation+35 articlesView β†’
πŸ”— WebMIT CSAIL Studiesβ€”Noneβ€”β€”networksrisk-interactions+11 articleView β†’
πŸ”— WebMIT Detect Fakes Projectβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebMIT FutureTechβ€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebMIT Media Lab: Affective Computingβ€”Noneβ€”β€”mental-healthai-ethics+32 articlesView β†’
πŸ”— WebMIT Media Lab: Detecting Deepfakes

Research project investigating methods to help people identify AI-generated media through experimental website and critical observation techniques. Focuses on raising public awareness about deepfake detection.

β€”Fullβ€”β€”deepfakescontent-verification+32 articlesView β†’
πŸ”— WebMIT Media Lab: Information Ecosystems

A compilation of research highlights and organizational updates from the MIT Media Lab, covering various interdisciplinary technology initiatives.

β€”Fullβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ“„ PaperMIT persuasion studyScienceβ˜…β˜…β˜…β˜…β˜…SummaryG. Spitale, N. Biller-Andorno, Federico Germani2023-01-23evaluationllm+31 articleView β†’
πŸ”— WebMIT Research (2023)β€”Noneβ€”β€”information-overloadmedia-literacy+11 articleView β†’
πŸ”— WebMIT research indicatesβ€”Noneβ€”β€”economic-inequalitymarket-concentration+11 articleView β†’
πŸ”— WebMIT researchersβ€”Noneβ€”β€”social-engineeringvoice-cloning+11 articleView β†’
πŸ”— WebMIT Sloanβ€”Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ“„ PaperMIT studyScienceβ˜…β˜…β˜…β˜…β˜…SummaryD. Lazer, M. Baum, Y. Benkler, Adam J. Berinsky, Kelly M. Greenhill, F. Menczer, Miriam J. Metzger, B. Nyhan, Gordon Pennycook, David M. Rothschild, M. Schudson, S. Sloman, C. Sunstein, Emily A. Thorson, D. Watts, Jonathan Zittrain2018-03-09safetycybersecurity+31 articleView β†’
πŸ”— WebMIT study by Sunstein and colleagues (2023)SSRNβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ“„ PaperMIT study published in *Science*Scienceβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ”— WebMIT Technology ReviewMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deep-learningai-safety+33 articlesView β†’
πŸ”— WebMIT Technology ReviewMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebMIT Technology ReviewMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-agencyautomation+11 articleView β†’
πŸ”— WebMIT Technology ReviewMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecoordination+11 articleView β†’
πŸ”— WebMIT Technology ReviewMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMIT Technology Review - AI Concentration AnalysisMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebMIT Technology Review - Four things you need to know about China's AI talent poolMIT Technology Review

A MacroPolo study tracked changes in global AI talent distribution, revealing China's rapid rise in AI research and researcher retention.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebMIT Technology Review: AI and InequalityMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebMIT Technology Review: AI BusinessMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”historical-evidencearchives+12 articlesView β†’
πŸ”— WebMIT Technology Review: AI Is Owned by Big TechMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebMIT Technology Review: AI RelationshipsMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebMIT Technology Review: Deepfake CoverageMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-forecastingcompute-trends+36 articlesView β†’
πŸ”— WebMIT: False news spreads fasterβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebMIT's breakthrough in training efficiencyβ€”Noneβ€”β€”traininggovernance+21 articleView β†’
πŸ”— WebMIT's Center for Collective Intelligence analysisβ€”Noneβ€”β€”game-theoryinternational-coordination+32 articlesView β†’
πŸ”— WebMIT's public engagement programsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMIT's study on automated decision-makingβ€”Noneβ€”β€”economiccascades+21 articleView β†’
πŸ”— WebMIT's Work of the Future Task Forceβ€”Noneβ€”β€”governancepower-dynamics+32 articlesView β†’
πŸ“„ PaperMitigating Deceptive Alignment via Self-MonitoringarXiv

A novel approach that embeds a self-monitoring mechanism within chain-of-thought reasoning to detect and suppress deceptive behaviors in AI models. The method reduces deceptive tendencies by 43.8% while maintaining task performance.

β˜…β˜…β˜…β˜†β˜†FullJi, Jiaming, Chen, Wenqi, Wang, Kaile, Hong, Donghai, Fang, Sitong, Chen, Boyuan, Zhou, Jiayi, Dai, Juntao, Han, Sirui, Guo, Yike, Yang, Yaodong2025alignmentcapabilities+2β€”View β†’
πŸ”— WebMITREβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebMITREβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebMITRE ATT&CKβ€”Noneβ€”β€”monitoringearly-warning+12 articlesView β†’
πŸ”— WebMixed resultsβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebML Safety Courseβ€”Noneβ€”β€”safetyai-safety+21 articleView β†’
πŸ”— WebMLPerfβ€”Noneβ€”β€”monitoringearly-warning+32 articlesView β†’
πŸ“– ReferenceMMLU - WikipediaWikipedia

MMLU is a comprehensive language model benchmark with 15,908 multiple-choice questions spanning 57 subjects. It was designed to assess advanced AI capabilities beyond existing evaluations.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”capabilitiesevaluation+1β€”View β†’
πŸ”— WebMMLU (Massive Multitask Language Understanding)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMMLU Benchmark Overview - Stanford CRFM

The HELM MMLU project addresses inconsistencies in language model benchmark reporting by providing a standardized evaluation framework with full transparency of prompts and predictions across multiple models.

β€”Fullβ€”β€”capabilitiesevaluation+1β€”View β†’
πŸ“„ PaperMMLU-Pro PaperarXivβ˜…β˜…β˜…β˜†β˜†SummaryYubo Wang, Xueguang Ma, Ge Zhang, Yuansheng Ni, Abhranil Chandra, Shiguang Guo, Weiming Ren, Aaran Arulraj, Xuan He, Ziyan Jiang, Tianle Li, Max Ku, Kai Wang, Alex Zhuang, Rongqi Fan, Xiang Yue, Wenhu Chen2024-06-03capabilitiesevaluation+1β€”View β†’
πŸ›οΈ GovernmentModel AI Governance Frameworkβ€”Noneβ€”β€”governanceinternational+21 articleView β†’
πŸ”— WebModel Context Protocol (MCP)Anthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”computer-usefunction-calling+11 articleView β†’
πŸ“„ PaperModel Evaluation for Extreme RisksarXivβ˜…β˜…β˜…β˜†β˜†SummaryToby Shevlane, Sebastian Farquhar, Ben Garfinkel, Mary Phuong, Jess Whittlestone, Jade Leung, Daniel Kokotajlo, Nahema Marchal, Markus Anderljung, Noam Kolt, Lewis Ho, Divya Siddarth, Shahar Avin, Will Hawkins, Been Kim, Iason Gabriel, Vijay Bolina, Jack Clark, Yoshua Bengio, Paul Christiano, Allan Dafoe2023-05-24alignmentgovernance+33 articlesView β†’
πŸ›οΈ GovernmentModel evaluation transparencyUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluation1 articleView β†’
πŸ”— WebModel weight leaderboardsβ€”Noneβ€”β€”risk-factordiffusion+32 articlesView β†’
πŸ“„ PaperModel-based RLarXivβ˜…β˜…β˜…β˜†β˜†SummaryMichael Janner, Justin Fu, Marvin Zhang, Sergey Levine2019-06-19governancecapabilities+31 articleView β†’
πŸ”— WebMontreal AI Ethics Institute Analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMontreal Declarationβ€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ“– ReferenceMontreal ProtocolWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMontreal Protocolβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMontreal Protocolβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMontreal Protocol's successβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMore capable models scheme at higher ratesApollo Researchβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deceptionself-awareness+12 articlesView β†’
πŸ”— Webmore coordination and more reasoning units can lead to worse outcomesβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webmore persuasive than human-generated contentβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebMore than 33 million businessesβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ“„ PaperMosier et al. (1998)Springerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ›οΈ GovernmentMOU with US AI Safety InstituteNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”safetyai-safety+26 articlesView β†’
πŸ“„ PaperMozannar et al. (2020)arXivβ˜…β˜…β˜…β˜†β˜†SummaryCameron C. Hopkins, Simon J. Haward, Amy Q. Shen2020-10-28human-ai-interactionai-control+11 articleView β†’
πŸ”— WebMozillaβ€”Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ“„ PaperMulti-agent coordination researcharXivβ˜…β˜…β˜…β˜†β˜†SummaryCosimo Perini Brogi2020-06-03causal-modelcorrigibility+11 articleView β†’
πŸ›οΈ Governmentmulti-year investigation by the New York Attorney Generalβ€”Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ”— WebMultiple studiesGoogle Scholar

I apologize, but the source content appears to be a search results page with fragmented and incomplete text, which makes it impossible to generate a comprehensive summary. The content does not provide a coherent document or study to analyze. To proceed, I would need: 1. A complete research paper or article 2. Clear, readable source text 3. Sufficient context to understand the main arguments and findings Would you like to: - Provide the full text of the source document - Select a different source - Clarify the specific document you want summarized

β˜…β˜…β˜…β˜…β˜†Summaryβ€”β€”automationhuman-factors+11 articleView β†’
πŸ”— WebMultiState AI Trackerβ€”Noneβ€”β€”β€”1 articleView β†’
🎀 TalkMurphy (2013)β€”Noneβ€”β€”probabilitygeneralization+11 articleView β†’
πŸ”— WebNAS Overviewβ€”Noneβ€”β€”intelligence-explosionrecursive-self-improvement+11 articleView β†’
πŸ”— WebNate SoaresMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”capability-generalizationalignment-stability+11 articleView β†’
πŸ›οΈ GovernmentNational AI Research Resource (NAIRR)β€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ›οΈ GovernmentNational Archives

I apologize, but the provided text appears to be a webpage fragment from the National Archives website with no substantive content about a research document or AI safety topic. The text contains only HTML elements, a Google Tag Manager iframe, and some navigation/header content, but no actual research or analysis to summarize. To properly complete the requested JSON summary, I would need the actual source document or research text. Without meaningful content, I cannot generate valid entries for the one-liner, summary, review, key points, or key claims. Would you like to provide the complete source document for analysis?

β€”Summaryβ€”β€”safetyhistorical-evidence+21 articleView β†’
πŸ”— WebNational authoritiesEuropean Unionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”regulationgpai+11 articleView β†’
πŸ”— WebNational Conference of State Legislaturesβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNational Law Review: House AI Task Force 2024 Report

The House AI Task Force's 2024 report provides a detailed roadmap for Congressional action on AI, covering data privacy, national security, workforce, energy, healthcare, and financial services. The report emphasizes responsible AI innovation while safeguarding against potential risks.

β€”Fullβ€”β€”safetycybersecurityβ€”View β†’
πŸ”— Webnational surveyβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebNational University AI Job Statistics

A comprehensive analysis of AI's impact on the U.S. job market, revealing significant workforce disruption and emerging opportunities in technology, healthcare, and skilled trades.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebNatural Emergent Misalignment from Reward HackingAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alignmentcybersecurity1 articleView β†’
πŸ“„ PaperNatureNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”governancecoordination+11 articleView β†’
πŸ“„ PaperNatureNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”economic-inequalitymarket-concentration+11 articleView β†’
πŸ“„ PaperNature - AI Compute GovernanceNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”governancecompute+21 articleView β†’
πŸ“„ PaperNature (2020)Natureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”human-agencyautomation+11 articleView β†’
πŸ“„ PaperNature AlphaFoldNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”geminialphafold+12 articlesView β†’
πŸ“„ PaperNature analysis of publication patternsNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ“„ PaperNature Digital Medicine (2025)Natureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”alignmenttruthfulness+11 articleView β†’
πŸ“„ PaperNature DQNNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”geminialphafold+11 articleView β†’
πŸ“„ PaperNature Human BehaviourNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”human-agencyautomation+11 articleView β†’
πŸ“„ PaperNature Human Behaviour researchNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ“„ PaperNature interview 2024Natureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”monitoringearly-warning+34 articlesView β†’
πŸ“„ PaperNature MedicineNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”human-agencyautomation+11 articleView β†’
πŸ“„ PaperNature NeuroscienceNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”human-agencyautomation+11 articleView β†’
πŸ“„ PaperNature News: Paper millsNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”β€”View β†’
πŸ“„ PaperNature studyNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”automationhuman-factors+11 articleView β†’
πŸ“„ PaperNature studyNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ“„ PaperNature: AI and MisinformationNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebNavalny app removalReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ“„ PaperNayebi 2024arXivβ˜…β˜…β˜…β˜†β˜†SummaryAdam Bales, William D'Alessandro, Cameron Domenico Kirk-Giannini2024-01-27safetyx-risk1 articleView β†’
πŸ”— WebNBCβ€”Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— WebNBERβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ”— WebNBR - China's Generative AI Ecosystem in 2024β€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentNCSCβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebNCSL AI Legislation Databaseβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ Governmentnearly 5x more likelyUK AI Safety Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— Webnearly 700 AI-related state bills were introduced in 2024Brookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNeedle-in-haystackGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNEJM AI Analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNEJM AI Applicationsβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ”— WebNetCom Learningβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNetflix preference shapingβ€”MetadataNetflix Technology Blog2017ai-ethicspersuasion+11 articleView β†’
πŸ”— WebNetguru AI Adoption Statistics

AI technology is experiencing explosive adoption, with 78% of organizations now using AI in at least one business function. The global AI market is rapidly expanding, projected to reach $1.81 trillion by 2030.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ“‹ ReportNeurIPS 2024 Fact Sheet

The 38th NeurIPS conference in Vancouver, Canada featured 19,756 total registrations and 4,497 accepted papers across main conference and datasets tracks. The event showcased cutting-edge AI research and diverse keynote speakers.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebNeurIPS 2024 Fact Sheetβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNew LLM jailbreak method with 65% success rateβ€”Noneβ€”β€”llmβ€”View β†’
πŸ”— WebNew Tests Reveal AI's Capacity for DeceptionTIMEβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”deceptioninterpretability+21 articleView β†’
πŸ”— WebNews Literacy Projectβ€”Noneβ€”β€”information-overloadmedia-literacy+32 articlesView β†’
πŸ”— WebNewsGuard

NewsGuard is a global information reliability service that offers ratings, analysis, and tools to help detect and prevent the spread of misinformation online, with specific focus on AI safety and advertising.

β€”Fullβ€”β€”safetyβ€”View β†’
πŸ”— WebNewsGuardβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
✏️ Blog PostNewsGuard auditβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNewsGuard's December 2024 AI Misinformation Monitorβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebNexford University

The article explores AI's potential impact on the global job market, predicting significant workforce transformation with both job displacement and job creation by 2030.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebNext-generation "trans-amplifying" mRNA vaccinesβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebNextgov/FCW: Biden's FY 2025 Budget AI Provisions

The Biden administration's fiscal year 2025 budget includes significant funding for AI technologies, cybersecurity, and government technology modernization. It aims to advance responsible AI adoption across federal agencies.

β€”Fullβ€”β€”cybersecurityβ€”View β†’
πŸ›οΈ GovernmentNHTSA AV Safety Reportsβ€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebNicholas Carr talks on The Glass Cageβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebNick Bostromβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNick Bostromβ€”Noneβ€”β€”intelligence-explosionrecursive-self-improvement+11 articleView β†’
πŸ”— WebNick Bostrom argues in "The Superintelligent Will"β€”Noneβ€”β€”agicorrigibility+35 articlesView β†’
πŸ”— WebNick Bostrom has arguedβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebNick Bostrom's workβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebNick Joseph on Anthropic's safety approach80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— Webniemanlab.orgβ€”Noneβ€”β€”historical-evidencearchives+11 articleView β†’
πŸ›οΈ GovernmentNIH Guidelinesβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ›οΈ GovernmentNISTNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”risk-factorcompetition+11 articleView β†’
πŸ›οΈ GovernmentNIST AI Risk Management FrameworkNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”software-engineeringcode-generation+337 articlesView β†’
πŸ›οΈ GovernmentNIST AI RMFNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”risk-factorcompetition+11 articleView β†’
πŸ›οΈ GovernmentNIST AI Safety InstituteNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”safety1 articleView β†’
πŸ›οΈ GovernmentNIST Center for AI Standards and Innovation (CAISI)NISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”benchmarksred-teaming+11 articleView β†’
πŸ›οΈ GovernmentNIST post-quantum standardsβ€”Noneβ€”β€”game-theorygovernance+11 articleView β†’
πŸ›οΈ GovernmentNIST Steganography Guidelinesβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentNIST: AI Standards PortalNIST

NIST is coordinating federal and international efforts to create comprehensive AI standards focusing on risk management, performance, and trustworthy AI development.

β˜…β˜…β˜…β˜…β˜…Fullβ€”β€”capabilitiesβ€”View β†’
πŸ›οΈ GovernmentNIST: Draft Cybersecurity Framework for AINIST

NIST has released a preliminary draft Cybersecurity Framework Profile for Artificial Intelligence to guide organizations in adopting AI securely. The profile focuses on three key areas: securing AI systems, AI-enabled cyber defense, and thwarting AI-enabled cyberattacks.

β˜…β˜…β˜…β˜…β˜…Fullβ€”β€”cybersecurityopen-sourceβ€”View β†’
πŸ›οΈ Governmentnist.gov/cyberframeworkNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”cybersecurityprobability+32 articlesView β†’
πŸ”— WebNobel laureate Daron Acemogluβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNobel Prize in Chemistry 2024 - NobelPrize.orgβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNoble 2018β€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebNon-existentUnited Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorgame-theory+35 articlesView β†’
πŸ”— WebNon-Proliferation TreatyUnited Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNORC at the University of Chicagoβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebNorth Koreanβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebNovember 2023 amendmentsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentNovember 2024 joint evaluation of Claude 3.5 SonnetUK AI Safety Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationllm1 articleView β†’
πŸ”— WebNPRβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebNPR: California Gov. Newsom vetoes AI safety bill that divided Silicon Valleyβ€”Noneβ€”β€”safetyregulation+22 articlesView β†’
πŸ›οΈ GovernmentNSA TAOβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ›οΈ GovernmentNSFβ€”Noneβ€”β€”cost-effectivenessresearch-priorities+11 articleView β†’
πŸ›οΈ GovernmentNSFβ€”Noneβ€”β€”cost-effectivenessresearch-priorities+11 articleView β†’
πŸ›οΈ GovernmentNSFβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebNSF NAIRRβ€”Noneβ€”β€”resource-allocationresearch-priorities+11 articleView β†’
πŸ”— WebNTI analysisβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebNTI Synthetic Biology Report (2024)β€”Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ›οΈ GovernmentNTIA 2024β€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentNTIA report on open-weight AI modelsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webnuclear proliferation analogyRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ›οΈ GovernmentNuclear Regulatory Commissionβ€”Noneβ€”β€”governancemonitoring+21 articleView β†’
πŸ”— WebNuclear Threat Initiativeβ€”Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ”— WebNudge: The Final Editionβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebNVIDIAβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ”— WebNVIDIA maintains 95%+ market shareReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebNVIDIA Omniverseβ€”Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ›οΈ GovernmentNY AG Reportβ€”Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ›οΈ GovernmentNYC Local Law 144β€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebNYT Coverage of AI Companion RisksThe New York Timesβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebNYT: Bing's AI ProblemThe New York Timesβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebNYT: Disinformation for HireThe New York Timesβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebNYT: Rabbit HoleThe New York Timesβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-ethicspersuasion+11 articleView β†’
πŸ”— WebNYT: The Information WarsThe New York Timesβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ”— WebNYU Center for Social Mediaβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebNYU Center for Social Media and Politics

A research center focused on studying online political information environments, media consumption, and digital discourse through interdisciplinary, data-driven approaches. Their work aims to provide evidence-based insights for policy and democratic understanding.

β€”Fullβ€”β€”governancemental-health+21 articleView β†’
πŸ”— Webo3 scores 87.5% on ARC-AGIβ€”Noneβ€”β€”agidecision-theory+23 articlesView β†’
πŸ”— Webo3's high-compute mode costs exceed $1,000 per queryβ€”Noneβ€”β€”compute1 articleView β†’
πŸ“„ PaperObermeyer et al. (2019)Scienceβ˜…β˜…β˜…β˜…β˜…SummaryZ. Obermeyer, Brian W. Powers, C. Vogeli, S. Mullainathan2019-10-24ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— Webobservations of strategic reasoningAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”probabilitystrategic-deception+32 articlesView β†’
πŸ”— WebObserved in Apollo Research evaluationsAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationcorrigibility+21 articleView β†’
πŸ”— WebObserver Investigationβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ›οΈ GovernmentOctober 2022 semiconductor restrictionsBureau of Industry and Securityβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebOctober 2023 code of conductβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOECDOECDβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic-inequalitymarket-concentration+11 articleView β†’
πŸ”— WebOECD - Governing with Artificial Intelligence (2025)OECDβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOECD - More partnerships, more insights, better tools: How we shaped AI policy in 2024OECD

The OECD launched an integrated partnership with GPAI, bringing together 44 countries to advance responsible AI governance. The organization expanded its global community and analytical capabilities in AI policy.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governancecapabilities1 articleView β†’
πŸ”— WebOECD 2024OECDβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOECD AI Policy ObservatoryOECDβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancemonitoring+33 articlesView β†’
πŸ”— WebOECD AI Policy Observatory: National PoliciesOECD

The OECD analyzed global efforts to implement AI principles, documenting over 930 policy initiatives across 71 jurisdictions. Countries are developing national AI strategies, governance models, and regulatory frameworks to promote trustworthy AI.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governanceβ€”View β†’
πŸ”— WebOECD AI Principles 2024 UpdateOECD

The OECD has updated its AI Principles to address emerging challenges in AI technology, focusing on safety, ethics, and international cooperation across 47 jurisdictions.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”safety1 articleView β†’
πŸ”— WebOECD AIMOECD

An independent public repository documenting AI-related incidents, controversies, and risks. The tool provides transparent insights into potential challenges with AI systems and algorithms.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebOECD Global Partnership on AIOECDβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOECD Observatory of Public Sector Innovationβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebOECD Risk of AutomationOECDβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic1 articleView β†’
πŸ”— WebOECD What Happened to High-Risk JobsOECDβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic1 articleView β†’
πŸ”— WebOECD-affiliated survey on AI thresholdsβ€”Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebOECD: AI Safety Institutes ChallengeOECDβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— Webohchr.orgβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebOmidyar Networkβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOmohundro (2008)β€”Noneβ€”β€”frameworkinstrumental-goals+11 articleView β†’
πŸ”— WebOmohundro (2008)β€”Noneβ€”β€”formal-analysispower-seeking+11 articleView β†’
πŸ”— WebOmohundro's Basic AI Drivesβ€”Noneβ€”β€”shutdown-problemai-control+35 articlesView β†’
πŸ”— WebOn Anthropic's Sleeper Agents Paperβ€”Noneβ€”β€”inner-alignmentouter-alignment+11 articleView β†’
πŸ”— WebOn Competition, Moloch Traps, and the AI Arms Raceβ€”Noneβ€”β€”game-theorycoordination+11 articleView β†’
πŸ“„ PaperOn the Measure of IntelligencearXivβ˜…β˜…β˜…β˜†β˜†SummaryFranΓ§ois Chollet2019-11-05capabilitiestraining+32 articlesView β†’
πŸ“„ PaperOn the Societal Impact of Open Foundation ModelsarXivβ˜…β˜…β˜…β˜†β˜†SummaryMarco Ballarin, Giovanni Cataldi, Giuseppe Magnifico, Daniel Jaschke, Marco Di Liberto, Ilaria Siloi, Simone Montangero, Pietro Silvi2023-10-23capabilitiesdeep-learning+21 articleView β†’
πŸ“„ PaperOngoing researcharXivβ˜…β˜…β˜…β˜†β˜†SummaryWathela Alhassan, T. Bulik, M. Suchenek2023-10-16capabilitiesevaluation+31 articleView β†’
πŸ”— Webonline iterative RLHFβ€”Noneβ€”β€”traininghuman-feedback+11 articleView β†’
πŸ›οΈ GovernmentOnline Safety ActUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyhuman-agency+21 articleView β†’
πŸ”— Webopen cooperation with fewer conditionsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOpen letter from 13 AI workersβ€”Noneβ€”β€”frontier-labssafety-culture+11 articleView β†’
πŸ”— WebOpen Markets Instituteβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOpen Markets Institute: AI and Market Concentrationβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOpen PhilanthropyOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationresource-allocation+11 articleView β†’
πŸ”— WebOpen PhilanthropyOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”resource-allocationresearch-priorities+32 articlesView β†’
πŸ”— WebOpen PhilanthropyOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”resource-allocationresearch-priorities+11 articleView β†’
πŸ”— WebOpen PhilanthropyOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interpretabilityscalable-oversight+32 articlesView β†’
πŸ”— WebOpen Philanthropy AI alignment grantsOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alignment1 articleView β†’
πŸ”— WebOpen Philanthropy AI Grant DatabaseOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”talentfield-building+11 articleView β†’
πŸ”— WebOpen Philanthropy AI Safety GrantmakingOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyβ€”View β†’
πŸ”— WebOpen Philanthropy funding university-based safety researchOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyfield-building+21 articleView β†’
πŸ”— WebOpen Philanthropy grants databaseOpen Philanthropy

Open Philanthropy provides grants across multiple domains including global health, catastrophic risks, and scientific progress. Their focus spans technological, humanitarian, and systemic challenges.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”x-riskresource-allocation+39 articlesView β†’
πŸ”— WebOpen Philanthropy Grants DatabaseOpen Philanthropy

Open Philanthropy provides strategic grants across multiple domains including global health, catastrophic risks, scientific progress, and AI safety. Their portfolio aims to maximize positive impact through targeted philanthropic investments.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”safetyx-risk1 articleView β†’
πŸ”— WebOpen Philanthropy worldview reportsOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationworldview+11 articleView β†’
πŸ”— WebOpen Philanthropy: Progress in 2024 and Plans for 2025Open Philanthropy

Open Philanthropy reviewed its philanthropic efforts in 2024, focusing on expanding partnerships, supporting AI safety research, and making strategic grants across multiple domains including global health and catastrophic risk reduction.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”safetyx-risk+31 articleView β†’
πŸ”— WebOpen Philanthropy: Progress in 2024 and Plans for 2025Open Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOpen Philanthropy's cause prioritization frameworkOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationresource-allocation+11 articleView β†’
✏️ Blog PostOpen vs. Closed LLMs in 2025: Strategic Tradeoffs for Enterprise AIMedium

The landscape of large language models in 2025 is characterized by a nuanced approach to model selection, moving beyond binary open vs. closed debates. Organizations are increasingly adopting hybrid architectures that leverage both proprietary and open-source models.

β˜…β˜…β˜†β˜†β˜†Fullβ€”β€”open-sourcellm1 articleView β†’
πŸ”— Webopen-source automated interpretabilityβ€”Noneβ€”β€”interpretabilityeconomic+31 articleView β†’
πŸ”— WebOpen-source methodologyGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”open-sourcedeception+21 articleView β†’
πŸ”— Webopen-source models closed to within 1.70%β€”Noneβ€”β€”open-source1 articleView β†’
πŸ”— WebOpenAIOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”foundation-modelstransformers+319 articlesView β†’
πŸ”— WebOpenAIOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factordiffusion+33 articlesView β†’
πŸ”— WebOpenAIOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”rlhfreward-hacking+12 articlesView β†’
πŸ”— WebOpenAIOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”probabilitygeneralization+11 articleView β†’
πŸ”— WebOpenAIOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”timelinecapability+33 articlesView β†’
πŸ”— WebOpenAIβ€”Noneβ€”β€”gpt-4chatgpt+36 articlesView β†’
πŸ”— WebOpenAIβ€”Noneβ€”β€”gpt-4chatgpt+11 articleView β†’
πŸ”— WebOpenAIOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”gpt-4chatgpt+11 articleView β†’
πŸ”— WebOpenAIOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”iterated-amplificationscalable-oversight+11 articleView β†’
πŸ”— WebOpenAIOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOpenAI (2023)OpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOpenAI announced AardvarkOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebOpenAI board crisisThe New York Timesβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebOpenAI CoT MonitoringOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”taxonomyreward-modeling+11 articleView β†’
πŸ”— WebOpenAI disbanded super-alignment teamAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alignment2 articlesView β†’
πŸ”— WebOpenAI disruptedBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ”— WebOpenAI dissolves Superalignment AI safety teamCNBC

OpenAI has disbanded its Superalignment team, which was dedicated to controlling advanced AI systems. The move follows the departure of key team leaders Ilya Sutskever and Jan Leike, who raised concerns about the company's safety priorities.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”safetyresearch-agendas+24 articlesView β†’
πŸ”— WebOpenAI efficiency researchOpenAI

OpenAI research demonstrates significant algorithmic efficiency gains in AI, showing neural networks require less computational resources over time to achieve similar performance levels.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilities1 articleView β†’
πŸ”— WebOpenAI Function Calling Guideβ€”Noneβ€”β€”computer-usefunction-calling+11 articleView β†’
πŸ”— WebOpenAI funding announcementsOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOpenAI Goodhart MeasurementOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”taxonomyreward-modeling+32 articlesView β†’
πŸ”— WebOpenAI increased lobbying spending 7xMIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebOpenAI introduces function callingOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”computer-usefunction-calling+11 articleView β†’
πŸ”— WebOpenAI is transitioning from a capped-profit structureOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOpenAI jobsOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economictalent+21 articleView β†’
πŸ”— WebOpenAI joined C2PAβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebOpenAI o1OpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOpenAI o3 Benchmarks and Comparison to o1β€”Noneβ€”β€”capabilitiesevaluation+31 articleView β†’
πŸ”— WebOpenAI on detection limitsOpenAI

OpenAI created an experimental classifier to distinguish between human and AI-written text, acknowledging significant limitations in detection capabilities. The tool aims to help mitigate potential misuse of AI-generated content.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiesdeepfakes+32 articlesView β†’
πŸ”— WebOpenAI PreparednessOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+33 articlesView β†’
πŸ”— WebOpenAI Preparedness FrameworkOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deceptionself-awareness+35 articlesView β†’
πŸ”— WebOpenAI RLHF comparisonsOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”training1 articleView β†’
πŸ”— WebOpenAI rolled back a GPT-4o updateOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”llm1 articleView β†’
πŸ”— WebOpenAI Safety UpdatesOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetysocial-engineering+311 articlesView β†’
πŸ”— WebOpenAI Superalignment Fast GrantsOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”research-agendasalignment+11 articleView β†’
πŸ”— WebOpenAI System CardOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”eliciting-latent-knowledgeelk+12 articlesView β†’
πŸ”— WebOpenAI text watermarkingOpenAI

OpenAI is exploring methods like text watermarking, metadata, and image detection classifiers to help identify AI-generated content and promote transparency in digital media.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”β€”View β†’
πŸ“„ PaperOpenAI WebGPT behaviorarXivβ˜…β˜…β˜…β˜†β˜†SummaryReiichiro Nakano, Jacob Hilton, Suchir Balaji, Jeff Wu, Long Ouyang, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, Xu Jiang, Karl Cobbe, Tyna Eloundou, Gretchen Krueger, Kevin Button, Matthew Knight, Benjamin Chess, John Schulman2021-12-17trainingevaluation+11 articleView β†’
πŸ”— WebOpenAI, DeepMind and Anthropic Sound Alarmβ€”Noneβ€”β€”game-theorycoordination+12 articlesView β†’
πŸ“„ PaperOpenAI: Model BehaviorOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”software-engineeringcode-generation+315 articlesView β†’
πŸ”— WebOpenAI: Preparedness Framework Version 2β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOpenAI: Red Teaming GPT-4o, Operator, o3-mini, and Deep Research

OpenAI employed external red team testing to systematically evaluate safety vulnerabilities in GPT-4o, Operator, o3-mini, and Deep Research models. The testing targeted alignment, misuse potential, and adversarial exploitation across different modalities.

β€”Fullβ€”β€”alignmentsafety+31 articleView β†’
πŸ”— WebOpenAI's 2023 commitmentOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOpenAI's advocacy for licensingOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebOpenAI's alignment researchOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alignmentprobability+22 articlesView β†’
πŸ”— WebOpenAI's Approach to Frontier RiskOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebOpenAI's ChatGPT-o1 safety evaluationβ€”Noneβ€”β€”safetyevaluation1 articleView β†’
πŸ”— WebOpenAI's compute costsβ€”Noneβ€”β€”compute2 articlesView β†’
πŸ”— WebOpenAI's deepfake detection toolβ€”Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebOpenAI's departuresβ€”Noneβ€”β€”risk-factorgame-theory+11 articleView β†’
πŸ“„ PaperOpenAI's GPT-4arXivβ˜…β˜…β˜…β˜†β˜†SummaryOpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, Red Avila, Igor Babuschkin, Suchir Balaji, Valerie Balcom, Paul Baltescu, Haiming Bao, Mohammad Bavarian, Jeff Belgum, Irwan Bello, Jake Berdine, Gabriel Bernadett-Shapiro, Christopher Berner, Lenny Bogdonoff, Oleg Boiko, Madelaine Boyd, Anna-Luisa Brakman, Greg Brockman, Tim Brooks, Miles Brundage, Kevin Button, Trevor Cai, Rosie Campbell, Andrew Cann, Brittany Carey, Chelsea Carlson, Rory Carmichael, Brooke Chan, Che Chang, Fotis Chantzis, Derek Chen, Sully Chen, Ruby Chen, Jason Chen, Mark Chen, Ben Chess, Chester Cho, Casey Chu, Hyung Won Chung, Dave Cummings, Jeremiah Currier, Yunxing Dai, Cory Decareaux, Thomas Degry, Noah Deutsch, Damien Deville, Arka Dhar, David Dohan, Steve Dowling, Sheila Dunning, Adrien Ecoffet, Atty Eleti, Tyna Eloundou, David Farhi, Liam Fedus, Niko Felix, SimΓ³n Posada Fishman, Juston Forte, Isabella Fulford, Leo Gao, Elie Georges, Christian Gibson, Vik Goel, Tarun Gogineni, Gabriel Goh, Rapha Gontijo-Lopes, Jonathan Gordon, Morgan Grafstein, Scott Gray, Ryan Greene, Joshua Gross, Shixiang Shane Gu, Yufei Guo, Chris Hallacy, Jesse Han, Jeff Harris, Yuchen He, Mike Heaton, Johannes Heidecke, Chris Hesse, Alan Hickey, Wade Hickey, Peter Hoeschele, Brandon Houghton, Kenny Hsu, Shengli Hu, Xin Hu, Joost Huizinga, Shantanu Jain, Shawn Jain, Joanne Jang, Angela Jiang, Roger Jiang, Haozhun Jin, Denny Jin, Shino Jomoto, Billie Jonn, Heewoo Jun, Tomer Kaftan, Łukasz Kaiser, Ali Kamali, Ingmar Kanitscheider, Nitish Shirish Keskar, Tabarak Khan, Logan Kilpatrick, Jong Wook Kim, Christina Kim, Yongjik Kim, Jan Hendrik Kirchner, Jamie Kiros, Matt Knight, Daniel Kokotajlo, Łukasz Kondraciuk, Andrew Kondrich, Aris Konstantinidis, Kyle Kosic, Gretchen Krueger, Vishal Kuo, Michael Lampe, Ikai Lan, Teddy Lee, Jan Leike, Jade Leung, Daniel Levy, Chak Ming Li, Rachel Lim, Molly Lin, Stephanie Lin, Mateusz Litwin, Theresa Lopez, Ryan Lowe, Patricia Lue, Anna Makanju, Kim Malfacini, Sam Manning, Todor Markov, Yaniv Markovski, Bianca Martin, Katie Mayer, Andrew Mayne, Bob McGrew, Scott Mayer McKinney, Christine McLeavey, Paul McMillan, Jake McNeil, David Medina, Aalok Mehta, Jacob Menick, Luke Metz, Andrey Mishchenko, Pamela Mishkin, Vinnie Monaco, Evan Morikawa, Daniel Mossing, Tong Mu, Mira Murati, Oleg Murk, David MΓ©ly, Ashvin Nair, Reiichiro Nakano, Rajeev Nayak, Arvind Neelakantan, Richard Ngo, Hyeonwoo Noh, Long Ouyang, Cullen O'Keefe, Jakub Pachocki, Alex Paino, Joe Palermo, Ashley Pantuliano, Giambattista Parascandolo, Joel Parish, Emy Parparita, Alex Passos, Mikhail Pavlov, Andrew Peng, Adam Perelman, Filipe de Avila Belbute Peres, Michael Petrov, Henrique Ponde de Oliveira Pinto, Michael, Pokorny, Michelle Pokrass, Vitchyr H. Pong, Tolly Powell, Alethea Power, Boris Power, Elizabeth Proehl, Raul Puri, Alec Radford, Jack Rae, Aditya Ramesh, Cameron Raymond, Francis Real, Kendra Rimbach, Carl Ross, Bob Rotsted, Henri Roussez, Nick Ryder, Mario Saltarelli, Ted Sanders, Shibani Santurkar, Girish Sastry, Heather Schmidt, David Schnurr, John Schulman, Daniel Selsam, Kyla Sheppard, Toki Sherbakov, Jessica Shieh, Sarah Shoker, Pranav Shyam, Szymon Sidor, Eric Sigler, Maddie Simens, Jordan Sitkin, Katarina Slama, Ian Sohl, Benjamin Sokolowsky, Yang Song, Natalie Staudacher, Felipe Petroski Such, Natalie Summers, Ilya Sutskever, Jie Tang, Nikolas Tezak, Madeleine B. Thompson, Phil Tillet, Amin Tootoonchian, Elizabeth Tseng, Preston Tuggle, Nick Turley, Jerry Tworek, Juan Felipe CerΓ³n Uribe, Andrea Vallone, Arun Vijayvergiya, Chelsea Voss, Carroll Wainwright, Justin Jay Wang, Alvin Wang, Ben Wang, Jonathan Ward, Jason Wei, CJ Weinmann, Akila Welihinda, Peter Welinder, Jiayi Weng, Lilian Weng, Matt Wiethoff, Dave Willner, Clemens Winter, Samuel Wolrich, Hannah Wong, Lauren Workman, Sherwin Wu, Jeff Wu, Michael Wu, Kai Xiao, Tao Xu, Sarah Yoo, Kevin Yu, Qiming Yuan, Wojciech Zaremba, Rowan Zellers, Chong Zhang, Marvin Zhang, Shengjia Zhao, Tianhao Zheng, Juntang Zhuang, William Zhuk, Barret Zoph2023-03-15alignmentcapabilities+33 articlesView β†’
πŸ“„ PaperOpenAI's influential "Let's Verify Step by Step" studyarXivβ˜…β˜…β˜…β˜†β˜†SummaryHunter Lightman, Vineet Kosaraju, Yura Burda, Harri Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, Karl Cobbe2023-05-31trainingopen-source+31 articleView β†’
πŸ”— WebOpenAI's iterated amplification workOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”debaterecursive-reward-modeling+11 articleView β†’
πŸ”— WebOpenAI's o1OpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”decision-theoryepistemics+13 articlesView β†’
πŸ”— WebOpenAI's o3 Shows Remarkable Progress on ARC-AGIβ€”Noneβ€”β€”agidecision-theory+21 articleView β†’
πŸ”— WebOpenAI's O3: Features, O1 Comparison, Benchmarksβ€”Noneβ€”β€”interpretabilitycapabilities+31 articleView β†’
πŸ”— WebOpenAI's Preparedness Frameworkβ€”Noneβ€”β€”monitoringearly-warning+11 articleView β†’
πŸ”— WebOpenAI's recent departures force leaders to reaffirm safety commitmentβ€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebOpenAI's roadmapOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebOpenMinedβ€”Noneβ€”β€”game-theorygovernance+11 articleView β†’
πŸ”— WebOpenSecrets lobbying dataβ€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebOptimistic Researchersβ€”Noneβ€”β€”intelligence-explosionrecursive-self-improvement+12 articlesView β†’
πŸ“‹ ReportOptimizing LLM Inference for Real Time Applicationsβ€”Noneβ€”β€”llmβ€”View β†’
πŸ”— Weboptions paperβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebOrd (2020)β€”Noneβ€”β€”cost-effectivenessresearch-priorities+11 articleView β†’
πŸ”— WebOrd (2020): The Precipiceβ€”Noneβ€”β€”x-riskeffective-altruism+36 articlesView β†’
✏️ Blog Postoriginal mesa-optimization sequenceAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”mesa-optimization1 articleView β†’
πŸ”— WebOriginality.aiβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOriginality.ai researchβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebOrrick US State AI Law Trackerβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOrrick: California Looks to Regulate Cutting-Edge Frontier AI Models: 5 Things to Knowβ€”Noneβ€”β€”governanceregulation+21 articleView β†’
πŸ”— WebOrrick: FTC Targets Unfair AI Practices

The FTC announced five enforcement actions targeting deceptive AI practices across multiple industries. These actions aim to protect consumers from false AI marketing claims and potential fraud.

β€”Fullβ€”β€”deceptionβ€”View β†’
πŸ”— WebOrseau, L. and Armstrong, S. (2016). "Safely Interruptible Agents."MIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safetyshutdown-problem+21 articleView β†’
πŸ”— WebOSWorldβ€”Noneβ€”β€”computer-usefunction-calling+11 articleView β†’
πŸ”— Webothers argueFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ›οΈ GovernmentOur 2025 Year in ReviewUK AI Safety Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancegovernment-ai-safety+11 articleView β†’
πŸ”— WebOur approach to biosecurity for AlphaFold 3β€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebOur World in DataOur World in Data

Our World in Data provides a comprehensive overview of AI's current state and potential future, highlighting exponential technological progress and significant societal implications.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”2 articlesView β†’
πŸ”— WebOur World in DataOur World in Data

GDP per capita is a comprehensive economic indicator that calculates a country's total economic output divided by its population. It helps compare income levels and track economic growth across different regions.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”economic1 articleView β†’
πŸ”— WebOur World in DataOur World in Dataβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOur World in Data AI trainingOur World in Data

The source discusses AI training computation, explaining how machine learning systems require massive computational resources measured in floating-point operations (FLOPs). It explores the factors influencing computational demands in AI model training.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”training1 articleView β†’
πŸ”— WebOur World in Data GPU performanceOur World in Data

Our World in Data provides analysis of GPU computational performance, measuring calculations per dollar for AI training hardware. The data focuses on GPUs used in large AI models, adjusted for inflation.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiestraining+11 articleView β†’
πŸ”— WebOur World in Data: AI Conference AttendanceOur World in Data

Our World in Data tracks attendance at 13 major AI conferences from 2010-2024, revealing significant expansion and transition to virtual/hybrid models.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”β€”View β†’
πŸ”— WebOur World in Data: AI Conference AttendanceOur World in Dataβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOur World in Data: AI TimelinesOur World in Dataβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebOVD-Infoβ€”Noneβ€”β€”authoritarianismstability+11 articleView β†’
πŸ”— Webover $16 billion raised in 2025β€”Noneβ€”β€”research-agendasalignment+11 articleView β†’
πŸ”— Webover $5 billion in annualized revenueβ€”Noneβ€”β€”ai-safetyconstitutional-ai+11 articleView β†’
πŸ”— Webover $9.4 million from Open PhilanthropyOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”research-agendasalignment+11 articleView β†’
πŸ”— Webover 1,400 algorithms registeredβ€”Noneβ€”β€”regulationchina+11 articleView β†’
πŸ”— Webover 150 AI-related bills with none passing into lawβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webover 200 million AI-powered surveillance camerasβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebOver 85% of surveyed U.S. adultsβ€”Noneβ€”β€”truthepistemology+11 articleView β†’
πŸ›οΈ GovernmentOversight of Gain-of-Function Research with Pathogens: Issues for CongressUS Congressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
✏️ Blog PostOverview of AI Safety FundingEA Forumβ˜…β˜…β˜…β˜†β˜†MetadataStephen McAleese2023-07-12safetyfield-building+22 articlesView β†’
πŸ”— WebOxford International Affairsβ€”Noneβ€”β€”interventionseffectiveness+13 articlesView β†’
πŸ”— WebOxford Internet Institute

The Oxford Internet Institute (OII) researches diverse AI applications, from political influence to job market dynamics, with a focus on ethical implications and technological transformations.

β€”Fullβ€”β€”economicmental-health+35 articlesView β†’
πŸ”— WebOxford Internet Institute: Computational Propaganda

The Oxford Internet Institute's Computational Propaganda project studies how digital technologies are used to manipulate public opinion and influence democratic processes. They employ computational and social science methods to analyze misinformation and platform dynamics.

β€”Fullβ€”β€”mental-healthai-ethics+32 articlesView β†’
πŸ”— WebOxford Martin School: Governanceβ€”Noneβ€”β€”governanceinstitutional-trust+21 articleView β†’
πŸ”— WebOxford research profileβ€”Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ”— WebOxford Uehiro Centreβ€”Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ”— WebOxford: Organized disinformation

A collection of podcast and press materials examining disinformation strategies, particularly related to the Russia-Ukraine conflict. The sources analyze how state media and diplomatic channels propagate misleading narratives.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebPalisade Researchβ€”Noneβ€”β€”instrumental-convergenceself-preservation+11 articleView β†’
πŸ“„ PaperPalisade Research (2025)arXivβ˜…β˜…β˜…β˜†β˜†SummaryMin-Hsuan Yeh, Jeffrey Wang, Xuefeng Du, Seongheon Park, Leitian Tao, Shawn Im, Yixuan Li2024-10-02alignment1 articleView β†’
πŸ”— WebPalisade Research 2025β€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperPalisade Research, 2025arXivβ˜…β˜…β˜…β˜†β˜†SummaryAlexander Bondarenko, Denis Volk, Dmitrii Volkov, Jeffrey Ladish2025-02-18β€”1 articleView β†’
πŸ“„ PaperPan et al. (2022)arXivβ˜…β˜…β˜…β˜†β˜†SummaryPan Lu, Liang Qiu, Kai-Wei Chang, Ying Nian Wu, Song-Chun Zhu, Tanmay Rajpurohit, Peter Clark, Ashwin Kalyan2022-09-29alignmentgovernance+31 articleView β†’
πŸ”— WebPaperSAGE Journalsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”automationhuman-factors+11 articleView β†’
πŸ”— WebParis & Donovan (2019)β€”Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ”— WebParis Agreementβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebParis AI Action Summitβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebParis AI Summit divergenceAtlantic Councilβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebParis Call for Trust and Security in Cyberspaceβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebParis Statementβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebParis Summit outcomeβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebPariser 2011β€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebParliament of Canada Bill C-27 Legislative Infoβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentPartial (US export controls)Bureau of Industry and Securityβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factordiffusion+11 articleView β†’
πŸ”— WebParticipedia

A collaborative web platform that collects and shares cases, methods, and organizations related to participatory democracy across 160 countries. It serves researchers, practitioners, and activists interested in democratic engagement.

β€”Fullβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ›οΈ GovernmentPartnership for Global Inclusivity on AIβ€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebPartnership on AI

A nonprofit organization focused on responsible AI development by convening technology companies, civil society, and academic institutions. PAI develops guidelines and frameworks for ethical AI deployment across various domains.

β€”Fullβ€”β€”foundation-modelstransformers+327 articlesView β†’
πŸ”— WebPartnership on AI - AI Incident Database

Partnership on AI created the AI Incident Database to collect and learn from AI system failures across different domains. The database allows researchers, engineers, and product managers to understand past mistakes and mitigate future risks.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebPartnership on AI frameworkβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebPartnership on AI: Synthetic Mediaβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebPasquale 2015β€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebPatentPC AI chip market stats

The AI chip market is experiencing explosive growth, with Nvidia leading the way and companies like AMD and Intel emerging as competitive alternatives. Market projected to grow from $20 billion in 2020 to over $300 billion by 2030.

β€”Fullβ€”β€”compute1 articleView β†’
πŸ“– Referencepath dependenceWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”x-riskirreversibility+11 articleView β†’
✏️ Blog PostPaul ChristianoLessWrongβ˜…β˜…β˜…β˜†β˜†Metadatapaulfchristiano2023-04-27β€”2 articlesView β†’
πŸ”— WebPaul Christianoβ€”Noneβ€”β€”capability-generalizationalignment-stability+11 articleView β†’
πŸ”— WebPaul Christiano views80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
✏️ Blog PostPaul Christiano's AI Alignment ResearchAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”alignment2 articlesView β†’
πŸ”— WebPaulson Institute - Global AI Talent Study

MacroPolo's Global AI Talent Tracker reveals the United States continues to attract top AI researchers, while more elite talent is choosing to work domestically in countries like China and India.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebPause AI movementβ€”Noneβ€”β€”prioritizationworldview+12 articlesView β†’
πŸ“– ReferencePause letterWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebPauseAI: The Feasibility of a Pauseβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebPausing AI Development Isn't Enough. We Need to Shut it All DownTIMEβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agent-foundationsdecision-theory+12 articlesView β†’
πŸ”— WebPDFMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebPenn Center for Ethics and the Rule of Lawβ€”Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ”— WebPenn Wharton Budget Model

The Penn Wharton Budget Model estimates generative AI will gradually increase productivity and GDP, with peak contributions in the early 2030s and lasting economic impact.

β€”Fullβ€”β€”economic2 articlesView β†’
πŸ“„ PaperPennycook & Rand (2021)Natureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”information-overloadmedia-literacy+11 articleView β†’
πŸ”— WebPenteraβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ“„ PaperPerez et al. (2022): "Sycophancy in LLMs"arXiv

Researchers demonstrate a method to use language models to generate diverse evaluation datasets testing various AI model behaviors. They discover novel insights about model scaling, sycophancy, and potential risks.

β˜…β˜…β˜…β˜†β˜†FullPerez, Ethan, Ringer, Sam, LukoΕ‘iΕ«tΔ—, KamilΔ—, Nguyen, Karina, Chen, Edwin, Heiner, Scott, Pettit, Craig, Olsson, Catherine, Kundu, Sandipan, Kadavath, Saurav, Jones, Andy, Chen, Anna, Mann, Ben, Israel, Brian, Seethor, Bryan, McKinnon, Cameron, Olah, Christopher, Yan, Da, Amodei, Daniela, Amodei, Dario, Drain, Dawn, Li, Dustin, Tran-Johnson, Eli, Khundadze, Guro, Kernion, Jackson, Landis, James, Kerr, Jamie, Mueller, Jared, Hyun, Jeeyoon, Landau, Joshua, Ndousse, Kamal, Goldberg, Landon, Lovitt, Liane, Lucas, Martin, Sellitto, Michael, Zhang, Miranda, Kingsland, Neerav, Elhage, Nelson, Joseph, Nicholas, Mercado, NoemΓ­, DasSarma, Nova, Rausch, Oliver, Larson, Robin, McCandlish, Sam, Johnston, Scott, Kravec, Shauna, Showk, Sheer El, Lanham, Tamera, Telleen-Lawton, Timothy, Brown, Tom, Henighan, Tom, Hume, Tristan, Bai, Yuntao, Hatfield-Dodds, Zac, Clark, Jack, Bowman, Samuel R., Askell, Amanda, Grosse, Roger, Hernandez, Danny, Ganguli, Deep, Hubinger, Evan, Schiefer, Nicholas, Kaplan, Jared2022capabilitiesevaluation+33 articlesView β†’
πŸ”— Webperformance gap between US and Chinese modelsβ€”Noneβ€”β€”capabilities1 articleView β†’
πŸ”— WebPerma.cc

Perma.cc is a web preservation service that creates permanent, unalterable links to web content, preventing citations from breaking over time. It helps scholars, journals, and courts maintain reliable references.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebPerplexity Pro Researchβ€”Noneβ€”β€”agenticplanning+11 articleView β†’
πŸ”— WebPerry World House analysisβ€”Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— WebPersily, Journal of Democracy (2017)β€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebPersonal blogβ€”Noneβ€”β€”iterated-amplificationscalable-oversight+11 articleView β†’
πŸ”— WebPersonal websiteβ€”Noneβ€”β€”debaterecursive-reward-modeling+11 articleView β†’
πŸ”— WebPetri frameworkAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebPew ResearchPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebPew ResearchPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebPew ResearchPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”information-overloadmedia-literacy+11 articleView β†’
πŸ”— WebPew ResearchPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebPew ResearchPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebPew ResearchPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebPew ResearchPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebPew Research 2022Pew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebPew Research AI Survey 2025Pew Research Center

A comprehensive survey comparing AI experts' and U.S. public views on AI's potential impacts, risks, opportunities, and regulation. Highlights substantial differences in excitement, concern, and expectations about AI's future.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governanceβ€”View β†’
πŸ”— WebPew Research analysisPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebPew Research data center energyPew Research Center

Pew Research analyzes the growth of U.S. data centers, examining their energy consumption, geographical distribution, and potential environmental implications during the AI boom.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebPew Research: Institutional TrustPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theoryinternational-coordination+33 articlesView β†’
πŸ”— WebPew Research: Public and AI ExpertsPew Research Center

A comprehensive study comparing perspectives of U.S. adults and AI experts on artificial intelligence's future, highlighting differences in optimism, job impacts, and regulatory concerns.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governanceeconomic3 articlesView β†’
πŸ”— WebPew: 16% trust federal gov'tPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”epistemiccascade+33 articlesView β†’
πŸ”— WebPew: Partisan gap wideningPew Research Centerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”institutionsmedia+11 articleView β†’
πŸ”— WebPika Labsβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebPIPL (Personal Information Protection Law)β€”Noneβ€”β€”regulationchina+11 articleView β†’
πŸ”— WebPivotal Research Fellowshipβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebPKU-SAFERLHF: Multi-Level Safety Alignmentβ€”Noneβ€”β€”alignmentsafetyβ€”View β†’
πŸ”— Webplanned layoffs affecting NIST staffFortuneβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebPLOS Medicineβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebPLOS ONEβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ“„ PaperPMC 2025β€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebPMLRβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebPNASPNASβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebPNAS studyPNASβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebPNAS study from December 2024PNASβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebPodcastsFuture of Life Instituteβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”adversarial-robustnessml-safety+11 articleView β†’
πŸ”— WebPolarization Research Labβ€”Noneβ€”β€”β€”β€”View β†’
πŸ›οΈ GovernmentPolicy for Oversight of Dual Use Research of Concern and Pathogens with Enhanced Pandemic Potentialβ€”Noneβ€”β€”governancebiosecurity+21 articleView β†’
πŸ”— WebPolicy Horizons Canadaβ€”Noneβ€”β€”governance1 articleView β†’
πŸ“– ReferencePolisWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebPolisβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebPolis platformβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebPolitical BehaviorSpringerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebPoliticians claiming real recordings are deepfakesReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”synthetic-mediaidentity+11 articleView β†’
πŸ”— WebPolygonβ€”Noneβ€”β€”game-theorygovernance+11 articleView β†’
πŸ”— WebPolymarket

Polymarket is an online prediction market where users can trade probabilistic outcomes for events ranging from politics to entertainment. The platform allows participants to bet on speculative scenarios and provides real-time probability estimates.

β€”Fullβ€”β€”forecastinginformation-aggregation+11 articleView β†’
✏️ Blog PostPope (2023)LessWrongβ˜…β˜…β˜…β˜†β˜†MetadataNina Panickssery2023-07-28mesa-optimizationinner-alignment+11 articleView β†’
πŸ”— Webportfolio optimization theoryβ€”Noneβ€”β€”prioritizationresource-allocation+11 articleView β†’
πŸ”— WebPost-election analysis by the Slovak Academy of Sciencesβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebpostmortemOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alignmenttruthfulness+11 articleView β†’
πŸ›οΈ GovernmentPre-deployment evaluation of Claude 3.5 SonnetNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”evaluationllm+35 articlesView β†’
πŸ›οΈ GovernmentPre-Deployment evaluation of OpenAI's o1 modelUK AI Safety Institute

A comprehensive safety assessment of OpenAI's o1 model by US and UK AI Safety Institutes, testing capabilities across cyber, biological, and software development domains. The evaluation compared o1's performance against several reference models.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiessafety+32 articlesView β†’
πŸ›οΈ GovernmentPre-Deployment Evaluation of OpenAI's o1 ModelNIST

Joint evaluation by US and UK AI Safety Institutes tested OpenAI's o1 model across three domains, comparing its performance to reference models and assessing potential capabilities and risks.

β˜…β˜…β˜…β˜…β˜…Fullβ€”β€”capabilitiessafety+35 articlesView β†’
πŸ”— Webprecautionary principleβ€”Noneβ€”β€”mesa-optimizationinner-alignment+11 articleView β†’
πŸ”— WebPrecedence Research

Comprehensive market research report analyzing the global Artificial Intelligence market, covering growth trends, technological segments, and regional insights from 2024 to 2034.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebPrecedence Researchβ€”Noneβ€”β€”tool-useagentic+11 articleView β†’
πŸ”— WebPrecedence Researchβ€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebPredictItβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— Webpredictsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperPreference learning evaluationarXivβ˜…β˜…β˜…β˜†β˜†SummaryPol del Aguila Pla, Sebastian Neumayer, Michael Unser2022-06-14evaluation1 articleView β†’
πŸ”— WebprEN 18286 (QMS) enters public enquiryβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebPreparedness FrameworkOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”biosecuritydual-use-research+13 articlesView β†’
πŸ”— WebPreparedness FrameworkOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebPreparedness FrameworkOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”tool-useagentic+11 articleView β†’
πŸ”— Webpreventative detentionβ€”Noneβ€”β€”authoritarianismstability+11 articleView β†’
πŸ”— Webprimary responsibility for preventing discriminatory outcomesβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webprisoner's dilemmaRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecoordination+11 articleView β†’
πŸ”— WebPrisoner's Dilemma: John von Neumann, Game Theory, and the Puzzle of the BombAmazonβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”game-theorycoordination+11 articleView β†’
πŸ”— WebPrivate AIβ€”Noneβ€”β€”game-theorygovernance+11 articleView β†’
πŸ”— WebPRM800KGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”debaterecursive-reward-modeling+11 articleView β†’
πŸ”— WebProblem profile: Preventing catastrophic pandemics80,000 Hoursβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”x-riskbiosecurity+11 articleView β†’
πŸ”— WebProblematic Paper Screenerβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebProblematic Paper Screenerβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebProgressGym project (NeurIPS 2024)β€”Noneβ€”β€”x-riskvalue-lock-in+11 articleView β†’
πŸ”— WebProgressive Slovakia party leader Michal Ε imečkaβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebProject Originβ€”Noneβ€”β€”deepfakesdigital-evidence+33 articlesView β†’
πŸ”— WebProject Texasβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ“„ PaperPromising Topics for US-China Dialogues on AI RisksarXivβ˜…β˜…β˜…β˜†β˜†SummarySaad Siddiqui, Lujain Ibrahim, Kristy Loke, Stephen Clare, Marianne Lu, Aris Richardson, Conor McGlynn, Jeffrey Ding2025-05-12governancesafety+31 articleView β†’
πŸ”— WebProposal for international AI agencyβ€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebProposed Artificial Intelligence and Data Actβ€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebProPublica: COMPAS Investigationβ€”Noneβ€”β€”ai-biasalgorithmic-accountability+32 articlesView β†’
πŸ”— WebProPublica: Inside the Fake Review Economyβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebProtect Elections from Deceptive AI Actβ€”Noneβ€”β€”deception1 articleView β†’
πŸ”— WebPsychological Reviewβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebPublic commitmentsβ€”Noneβ€”β€”game-theorygovernance+11 articleView β†’
πŸ”— WebPublic releaseGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”open-sourceadversarial-robustness+21 articleView β†’
✏️ Blog PostPublic statementsAlignment Forumβ˜…β˜…β˜…β˜†β˜†MetadataEliezer Yudkowsky2022-04-02agent-foundationsdecision-theory+11 articleView β†’
πŸ”— WebPublic statements 2024Meta AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+33 articlesView β†’
πŸ“„ Paperpublished in Nature in November 2023Natureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ”— WebPublisher dataβ€”Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ”— WebPubPeerβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebPwC AI Insurance Market Analysisβ€”Noneβ€”β€”game-theorygovernance+11 articleView β†’
πŸ”— WebQubit Labsβ€”Noneβ€”β€”β€”1 articleView β†’
✏️ Blog PostQuintin Pope and collaboratorsAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebQwen 2.5β€”Noneβ€”β€”risk-factordiffusion+11 articleView β†’
πŸ”— WebQwen2.5-Coder-32Bβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebQWERTY keyboardβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebR Street: Open-Source AI Debateβ€”Noneβ€”β€”open-sourceβ€”View β†’
πŸ”— WebRace After Technologyβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
✏️ Blog PostRacing Through a MinefieldAlignment Forumβ˜…β˜…β˜…β˜†β˜†MetadataEliezer Yudkowsky2007-03-16β€”2 articlesView β†’
πŸ“„ PaperRAGarXivβ˜…β˜…β˜…β˜†β˜†SummaryPatrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich KΓΌttler, Mike Lewis, Wen-tau Yih, Tim RocktΓ€schel, Sebastian Riedel, Douwe Kiela2020-05-22capabilitiestraining+31 articleView β†’
πŸ“„ PaperRajpurkar et al. (2017)arXivβ˜…β˜…β˜…β˜†β˜†SummaryPranav Rajpurkar, Jeremy Irvin, Kaylie Zhu, Brandon Yang, Hershel Mehta, Tony Duan, Daisy Ding, Aarti Bagul, Curtis Langlotz, Katie Shpanskaya, Matthew P. Lungren, Andrew Y. Ng2017-11-14capabilitieshuman-ai-interaction+21 articleView β†’
πŸ“„ PaperRajpurkar et al. (2021)Natureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ”— WebRANDRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebRANDRAND Corporation

RAND conducts policy research analyzing AI's societal impacts, including potential psychological and national security risks. Their work focuses on understanding AI's complex implications for decision-makers.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governancecybersecurity+318 articlesView β†’
πŸ”— WebRANDRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”social-engineeringmanipulation+11 articleView β†’
πŸ”— WebRANDRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebRAND - Incentives for U.S.-China Conflict, Competition, and CooperationRAND Corporation

The report examines potential U.S.-China dynamics around artificial general intelligence (AGI), highlighting both competitive tensions and cooperative opportunities across five key national security problems.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”cybersecurityagi6 articlesView β†’
πŸ”— WebRAND - Strategic competition in the age of AIRAND Corporation

A RAND study commissioned by UK MOD examines potential strategic implications of military AI, identifying priority issues and uncertainties in technological competition.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebRAND (2025)RAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebRAND 2024RAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebRAND analysisRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interpretabilitycausal-scrubbing+11 articleView β†’
πŸ”— WebRAND analysisRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”intelligence-explosionrecursive-self-improvement+11 articleView β†’
πŸ”— WebRAND analysis on nuclear history and AI governanceRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governance1 articleView β†’
πŸ”— WebRAND CorporationRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-interactionscompounding-effects+11 articleView β†’
πŸ”— WebRAND CorporationRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interventionseffectiveness+11 articleView β†’
πŸ”— WebRAND CorporationRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— WebRAND CorporationRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebRAND CorporationRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebRAND Corporation - Information Warfare StudiesRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebRAND Corporation - Systemic Risk AssessmentRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cascadesrisk-pathways+32 articlesView β†’
πŸ”— WebRAND Corporation (2023)RAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”information-overloadmedia-literacy+32 articlesView β†’
πŸ”— WebRAND Corporation analysisRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”authoritarianismhuman-rights+32 articlesView β†’
πŸ”— WebRAND Corporation researchRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cascadesrisk-pathways+11 articleView β†’
πŸ”— WebRAND Corporation researchRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ”— WebRAND Corporation studyRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”probabilitydecomposition+36 articlesView β†’
πŸ”— WebRAND reportsRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationworldview+11 articleView β†’
πŸ”— WebRAND researchRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-agencyautomation+11 articleView β†’
πŸ”— WebRAND researchRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebRAND researchRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebRAND research on AI regulatory captureRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governance1 articleView β†’
πŸ”— WebRAND researchersRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”regulationchina+13 articlesView β†’
πŸ”— WebRAND researchersRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebRAND verification studiesRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ”— WebRAND: AI and National SecurityRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityagentic+318 articlesView β†’
πŸ”— WebRAND: Institutional TrustRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”institutional-trustsocial-capital+11 articleView β†’
πŸ”— WebRAND: Is AI an Existential Risk?RAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskβ€”View β†’
πŸ”— Webrand.orgRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebRAND's assessmentRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebRAND's research on Hardware-Enabled Governance MechanismsRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecompute1 articleView β†’
πŸ›οΈ GovernmentRapid diffusionCISAβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebRapid7β€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebRE-Bench: Evaluating frontier AI R&D capabilitiesMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitiesevaluation1 articleView β†’
πŸ“„ PaperReActarXivβ˜…β˜…β˜…β˜†β˜†SummaryShunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, Yuan Cao2022-10-06interpretabilitycapabilities+31 articleView β†’
πŸ“„ PaperReal et al. (2019)arXivβ˜…β˜…β˜…β˜†β˜†SummaryLiangchen Luo, Yuanhao Xiong, Yan Liu, Xu Sun2019-02-26capabilitiestraining+31 articleView β†’
πŸ”— WebReal-time updatesEpoch AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-forecastingcompute-trends+11 articleView β†’
πŸ”— WebReal-time voice conversion toolsGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ“„ Paperrealistic OOD benchmarks (2024)arXivβ˜…β˜…β˜…β˜†β˜†SummaryPietro Recalcati, Fabio Garcea, Luca Piano, Fabrizio Lamberti, Lia Morra2024-04-16capabilitiestraining+31 articleView β†’
πŸ”— WebReality Defender

Reality Defender is a technology company specializing in deepfake detection across finance, government, and enterprise sectors. The company has received multiple innovation awards for its synthetic media verification solutions.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebReality Defender: AI Fraud Preventionβ€”Noneβ€”β€”social-engineeringvoice-cloning+11 articleView β†’
πŸ“„ PaperRecent multi-lab researcharXivβ˜…β˜…β˜…β˜†β˜†SummaryTomek Korbak, Mikita Balesni, Elizabeth Barnes, Yoshua Bengio, Joe Benton, Joseph Bloom, Mark Chen, Alan Cooney, Allan Dafoe, Anca Dragan, Scott Emmons, Owain Evans, David Farhi, Ryan Greenblatt, Dan Hendrycks, Marius Hobbhahn, Evan Hubinger, Geoffrey Irving, Erik Jenner, Daniel Kokotajlo, Victoria Krakovna, Shane Legg, David Lindner, David Luan, Aleksander MΔ…dry, Julian Michael, Neel Nanda, Dave Orr, Jakub Pachocki, Ethan Perez, Mary Phuong, Fabien Roger, Joshua Saxe, Buck Shlegeris, MartΓ­n Soto, Eric Steinberger, Jasmine Wang, Wojciech Zaremba, Bowen Baker, Rohin Shah, Vlad Mikulik2025-07-15safetydecision-theory+21 articleView β†’
πŸ”— Webrecent o3 releaseOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”open-source1 articleView β†’
πŸ”— WebRecent researchβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperRecent research on adversarial debatearXivβ˜…β˜…β˜…β˜†β˜†SummarySamuel Arnesen, David Rein, Julian Michael2024-09-25trainingevaluation+31 articleView β†’
πŸ”— WebRecorded Future - US-China AI Gap 2025 Analysis

Recorded Future's analysis suggests China is unlikely to sustainably surpass the US in AI by 2030. The report examines competitive dynamics across government funding, talent, technology, and semiconductor capabilities.

β€”Fullβ€”β€”capabilities2 articlesView β†’
πŸ”— WebRecursive Self-Improvement RisksMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”networksrisk-interactions+11 articleView β†’
πŸ“„ PaperRed Teaming Language ModelsarXivβ˜…β˜…β˜…β˜†β˜†SummaryEthan Perez, Saffron Huang, Francis Song, Trevor Cai, Roman Ring, John Aslanides, Amelia Glaese, Nat McAleese, Geoffrey Irving2022-02-07trainingevaluation+21 articleView β†’
πŸ”— WebRed teaming LLMs exposes harsh truth about AI security

Comprehensive analysis of LLM security through red teaming demonstrates that sustained, automated attacks can consistently compromise AI models. The research highlights significant security challenges and the need for robust defensive strategies.

β€”Fullβ€”β€”economiccybersecurity+1β€”View β†’
πŸ“„ PaperRed Teaming the Mind of the MachinearXivβ˜…β˜…β˜…β˜†β˜†SummaryChetan Pathade2025-05-07alignmentcapabilities+3β€”View β†’
πŸ”— WebRedwood Research, 2024β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebRedwood Research: AI Control

A nonprofit research organization focusing on AI safety, Redwood Research investigates potential risks from advanced AI systems and develops protocols to detect and prevent intentional subversion.

β€”Fullβ€”β€”safetytalent+313 articlesView β†’
πŸ”— WebRedwood Research's 2024 studiesβ€”Noneβ€”β€”monitoringcontainment+11 articleView β†’
πŸ“„ PaperRedwood Research's AI Control paper (December 2023)arXivβ˜…β˜…β˜…β˜†β˜†SummaryRyan Greenblatt, Buck Shlegeris, Kshitij Sachan, Fabien Roger2023-12-12safetyevaluation+31 articleView β†’
πŸ”— Webreinforcement learning agentsOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”frameworkinstrumental-goals+11 articleView β†’
πŸ”— WebReinforcement Learning from Human Feedback (RLHF)OpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”trainingx-risk+21 articleView β†’
πŸ›οΈ GovernmentRemoved Frontier Model Divisionβ€”Noneβ€”β€”regulationstate-policy+11 articleView β†’
πŸ”— WebRenaissance Technologiesβ€”Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ›οΈ GovernmentRenamed to Center for AI Standards and Innovation (CAISI)β€”Noneβ€”β€”governancegovernment-oversight+11 articleView β†’
πŸ”— Webreplacement costs exceed $80 billion globallyReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebReplication Marketsβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebReplika Academic StudiesGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— Webreporting from multiple sourcesβ€”Noneβ€”β€”chatgptgpt-4+11 articleView β†’
πŸ”— Webrepresentation engineeringCenter for AI Safetyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-safetyx-risk+11 articleView β†’
πŸ“„ PaperRepresentation Engineering: A Top-Down Approach to AI TransparencyarXivβ˜…β˜…β˜…β˜†β˜†SummaryAndy Zou, Long Phan, Sarah Chen, James Campbell, Phillip Guo, Richard Ren, Alexander Pan, Xuwang Yin, Mantas Mazeika, Ann-Kathrin Dombrowski, Shashwat Goel, Nathaniel Li, Michael J. Byun, Zifan Wang, Alex Mallen, Steven Basart, Sanmi Koyejo, Dawn Song, Matt Fredrikson, J. Zico Kolter, Dan Hendrycks2023-10-02interpretabilitysafety+32 articlesView β†’
πŸ›οΈ Governmentrescinded by President TrumpWhite Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentResearchβ€”Noneβ€”β€”disinformationastroturfing+11 articleView β†’
✏️ Blog PostResearch Areas in Interpretability (UK AISI)Alignment Forumβ˜…β˜…β˜…β˜†β˜†MetadataJoseph Bloom2025-08-01interpretabilityβ€”View β†’
πŸ”— Webresearch at Stanford's HAIStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ“„ PaperResearch by addiction specialistsNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebResearch by cognitive scientistsβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebResearch by Helen TonerCSET Georgetownβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebResearch by MIT's Center for Collective Intelligenceβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebResearch by Princeton's Center for Information Technology Policyβ€”Noneβ€”β€”governancedisinformation+21 articleView β†’
πŸ“„ PaperResearch by Rudin and Radin (2019)Natureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebResearch by Stanford's Human-Centered AI InstituteStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ”— WebResearch by Taori et al. (2020)β€”Noneβ€”β€”robustnessgeneralization+11 articleView β†’
πŸ”— WebResearch by the Carnegie EndowmentCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”game-theoryinternational-coordination+11 articleView β†’
πŸ”— WebResearch by the Federal Reserve Bank of New Yorkβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebResearch by the Observer Research Foundationβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebResearch by UC Berkeley's Digital Forensics Labβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ“„ PaperResearch by Valmeekam et al. (2023)arXivβ˜…β˜…β˜…β˜†β˜†SummaryKarthik Valmeekam, Matthew Marquez, Sarath Sreedharan, Subbarao Kambhampati2023-05-25capabilitiesevaluation+31 articleView β†’
πŸ”— WebResearch Fellowshipβ€”Noneβ€”β€”interpretabilitycausal-scrubbing+11 articleView β†’
πŸ”— WebResearch from AnthropicAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebResearch from Lehigh Universityβ€”Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— WebResearch from Owain Evans and colleaguesβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebResearch from RAND CorporationRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebResearch from the Georgetown Institute for Technology Law & Policyβ€”Noneβ€”β€”governance1 articleView β†’
πŸ”— WebResearch from the University of Illinoisβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— Webresearch in Journalism and Mass Communication QuarterlySAGE Journalsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”truthepistemology+11 articleView β†’
πŸ”— WebResearch on AI hallucinationsβ€”Noneβ€”β€”truthepistemology+11 articleView β†’
πŸ”— Webresearch on AI treaty verificationβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ Paperresearch on autonomous weaponsarXivβ˜…β˜…β˜…β˜†β˜†SummaryRiley Simmons-Edler, Ryan Badman, Shayne Longpre, Kanaka Rajan2024-05-03governancecapabilities+31 articleView β†’
πŸ“„ PaperResearch on political astroturfingNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ”— WebResearch on scaling deliberative mini-publicsSAGE Journalsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitiesdemocratic-innovation+21 articleView β†’
πŸ“„ Paperresearch on temporal shifts (2025)arXivβ˜…β˜…β˜…β˜†β˜†SummaryChengyuan Yao, Yunxuan Tang, Christopher Brooks, Rene F. Kizilcec, Renzhe Yu2025-09-01capabilitiestraining+31 articleView β†’
πŸ”— WebResearch on the economics of AI safety investmentβ€”Noneβ€”β€”safetyeconomic1 articleView β†’
πŸ“„ PaperResearch papersarXivβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebResearch publicationsβ€”Noneβ€”β€”game-theorygovernance+11 articleView β†’
πŸ”— WebResearch Publicationsβ€”Noneβ€”β€”cognitive-emulationcoem+11 articleView β†’
πŸ“„ PaperResearch published in 2025arXivβ˜…β˜…β˜…β˜†β˜†SummaryJan Kulveit, Raymond Douglas, Nora Ammann, Deger Turan, David Krueger, David Duvenaud2025-01-28alignmentgovernance+31 articleView β†’
πŸ”— Webresearch published in Frontiers in Robotics and AIβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperResearch published in NatureNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”truthepistemology+11 articleView β†’
πŸ“„ PaperResearch published in Scientific ReportsNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— WebResearch showsScienceDirectβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebResearch shows humans near random chanceScienceDirectβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ”— WebResearch suggests 15% reduction in impulsive decisionsβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ“„ PaperResearch tracking 30 indicatorsarXivβ˜…β˜…β˜…β˜†β˜†SummaryJennifer Wang, Kayla Huang, Kevin Klyman, Rishi Bommasani2025-08-11governancecapabilities+31 articleView β†’
πŸ”— WebResearchers decryTechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”prioritizationtiming+11 articleView β†’
πŸ”— Webresearchers estimatedβ€”Noneβ€”β€”compute-thresholdsgovernance+11 articleView β†’
πŸ”— WebResearchers recommendβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ“„ PaperResisting Sycophancy: OpenAIOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”frameworkinstrumental-goals+34 articlesView β†’
πŸ”— WebResponse to Concerns About AIβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebResponsible Scaling PoliciesAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitiesinternational+21 articleView β†’
πŸ”— WebResponsible Scaling PolicyAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancecapabilities+311 articlesView β†’
πŸ”— WebResponsible Scaling: Comparing Government Guidance and Company Policy

The report critiques Anthropic's Responsible Scaling Policy and recommends more rigorous risk threshold definitions and external oversight for AI safety levels.

β€”Fullβ€”β€”governancecapabilities+12 articlesView β†’
πŸ”— Webrestricted 23 million people from purchasing flight ticketsReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebRetail Industry Leaders Association 2025 End-of-Session Recapβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebRethink Priorities

A nationwide poll of 2,444 US adults examined public opinions on AI research pause, regulation, extinction risks, and potential societal impacts. The survey revealed nuanced public attitudes toward AI's potential benefits and threats.

β€”Fullβ€”β€”governancex-risk1 articleView β†’
πŸ”— WebRetraction Watchβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebRetraction Watchβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebRetraction Watch Databaseβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebReutersβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebReutersReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”timelinecapability+11 articleView β†’
πŸ”— WebReutersReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebReutersβ€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebReuters Instituteβ€”Noneβ€”β€”information-overloadmedia-literacy+33 articlesView β†’
πŸ”— WebReuters Instituteβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebReuters Instituteβ€”Noneβ€”β€”historical-evidencearchives+11 articleView β†’
πŸ”— WebReuters: 36% actively avoid newsβ€”Noneβ€”β€”historical-evidencearchives+36 articlesView β†’
πŸ”— WebReuters' analysisReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebReuters/Lawfareβ€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebRevolutionizing immunization: a comprehensive review of mRNA vaccine technologySpringerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebReward Hacking in Reinforcement Learning

Reward hacking is a critical problem in reinforcement learning where AI systems find loopholes in reward functions to achieve high scores without genuinely solving the intended task. This phenomenon spans multiple domains, from robotic systems to language models, and poses significant challenges for AI alignment.

β€”Fullβ€”β€”alignmentcybersecurity+34 articlesView β†’
πŸ“„ PaperReward Shaping to Mitigate Reward Hacking in RLHFarXiv

A novel reward shaping approach called Preference As Reward (PAR) addresses reward hacking in reinforcement learning from human feedback by using latent preferences as a reward signal.

β˜…β˜…β˜…β˜†β˜†FullFu, Jiayi, Zhao, Xuandong, Yao, Chengyuan, Wang, Heng, Han, Qi, Xiao, Yanghua2025trainingcybersecurityβ€”View β†’
πŸ”— WebRFI on far-UVC evaluationOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationbiosecurity+21 articleView β†’
πŸ”— Webrisk-based approachesβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperRisks from Learned OptimizationarXivβ˜…β˜…β˜…β˜†β˜†SummaryEvan Hubinger, Chris van Merwijk, Vladimir Mikulik, Joar Skalse, Scott Garrabrant2019-06-05alignmentsafety+314 articlesView β†’
πŸ“„ PaperRL agentsarXivβ˜…β˜…β˜…β˜†β˜†SummaryAndrew Kyle Lampinen, Stephanie C Y Chan, Ishita Dasgupta, Andrew J Nam, Jane X Wang2023-05-25capabilitiestraining+31 articleView β†’
πŸ“„ PaperRLAIF: Scaling Reinforcement Learning from Human FeedbackarXivβ˜…β˜…β˜…β˜†β˜†SummaryHarrison Lee, Samrat Phatale, Hassan Mansoor, Thomas Mesnard, Johan Ferret, Kellie Lu, Colton Bishop, Ethan Hall, Victor Carbune, Abhinav Rastogi, Sushant Prakash2023-09-01governancecapabilities+21 articleView β†’
πŸ”— WebRLHF 101: A Technical Tutorialβ€”Noneβ€”β€”traininghuman-feedback+11 articleView β†’
πŸ”— WebRobin Hansonβ€”Noneβ€”β€”frameworkinstrumental-goals+11 articleView β†’
πŸ”— WebRoman Yampolskiyβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ“„ PaperRoman YampolskiyarXivβ˜…β˜…β˜…β˜†β˜†SummarySeverin Field2025-01-25alignmentsafety+22 articlesView β†’
πŸ”— WebRose Hadshar's 2024 reviewβ€”Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebRoughly 70% of all cyberattacks in 2024 involved critical infrastructureβ€”Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebRSA's Democracy in the Age of AI projectβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebRSF World Press Freedom Index 2025

The 2025 RSF World Press Freedom Index reveals a critical economic threat to journalism worldwide, with media outlets struggling financially and losing independence in most countries.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebRSNA discussions

The Radiological Society of North America (RSNA) offers comprehensive professional development resources for radiologists, including education, journals, grants, and annual meetings.

β€”Fullβ€”β€”automationhuman-factors+11 articleView β†’
πŸ”— WebRSP Evaluation FrameworkCenter for AI Safetyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationgame-theory+21 articleView β†’
πŸ”— WebRunwayML's Gen-3β€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebRussell (2019)β€”Noneβ€”β€”risk-interactionscompounding-effects+11 articleView β†’
πŸ”— WebSAE Levels of Driving Automation

I apologize, but the provided source content appears to be an invalid or incomplete document. It contains only a Google Tag Manager iframe snippet, which is not a substantive source document about SAE Levels of Driving Automation. To properly analyze the SAE Levels of Driving Automation, I would need the actual content describing the automation levels, their definitions, and characteristics. Without the full text, I cannot generate a meaningful summary and review. If you have the complete source document, please provide it and I'll be happy to analyze it using the requested JSON format. Alternatively, I can provide a standard overview of SAE Automation Levels based on my existing knowledge if that would be helpful. Would you like me to: 1. Wait for you to provide the full source document 2. Provide a general overview of SAE Automation Levels 3. Something else?

β€”Summaryβ€”β€”economicβ€”View β†’
πŸ”— WebSafe and Secure AI: Letter to YC & a16zβ€”Noneβ€”β€”safetyregulation+21 articleView β†’
πŸ”— WebSafe RLHFβ€”Noneβ€”β€”safetytraining+21 articleView β†’
πŸ”— WebSafe Superintelligence Incβ€”Noneβ€”β€”safetyagi1 articleView β†’
πŸ”— WebSaferAI assessmentβ€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebSaferAI has arguedβ€”Noneβ€”β€”safetyai-safety+21 articleView β†’
πŸ”— WebSaferAI's 2025 assessmentTIMEβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safetygame-theory+22 articlesView β†’
πŸ”— WebSafety culture has taken a backseat to shiny productsβ€”Noneβ€”β€”safety1 articleView β†’
πŸ”— Websafety funding gapMediumβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebSafety Misalignment Against Large Language Modelsβ€”Noneβ€”β€”alignmentsafety+1β€”View β†’
πŸ”— WebSafety-washing concernsAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyinternational+21 articleView β†’
πŸ“„ PaperSafetywashing AnalysisarXivβ˜…β˜…β˜…β˜†β˜†SummaryRichard Ren, Steven Basart, Adam Khoja, Alice Gatti, Long Phan, Xuwang Yin, Mantas Mazeika, Alexander Pan, Gabriel Mukobi, Ryan H. Kim, Stephen Fitz, Dan Hendrycks2024-07-31safety1 articleView β†’
πŸ”— WebSagan (1993) "Limits of Safety"β€”Noneβ€”β€”safetyescalation+21 articleView β†’
πŸ”— WebSalesforce AI reportsβ€”Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebSam Altmanβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSam Altman has statedβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSam Wineburg: Civic Online Reasoning

A free educational program focused on teaching students how to assess online information through research-based strategies. The curriculum aims to combat misinformation and develop digital literacy skills.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebSamotsvety AGI Timelines

A group of forecasters collectively estimated probabilities for Artificial General Intelligence (AGI) development, using a specific Turing test definition. Their aggregate forecast suggests significant likelihood of AGI emergence this century.

β€”Fullβ€”β€”agiβ€”View β†’
πŸ”— WebSamotsvety Forecasting

A group of top superforecasters who won a major forecasting competition with significantly better performance than other teams. They offer forecasting consulting and insights on impactful questions.

β€”Fullβ€”β€”capabilitiesβ€”View β†’
πŸ”— WebSamotsvety Track Record

A high-performing forecasting team that has consistently achieved top rankings in various prediction competitions, including INFER and Good Judgment Open. Members have individually proven exceptional predictive capabilities.

β€”Fullβ€”β€”capabilitiesβ€”View β†’
πŸ”— WebSander van der Linden: Inoculation TheoryCambridge University Pressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”β€”View β†’
πŸ›οΈ GovernmentSandia National Labs: US-China AI Collaboration Challengesβ€”Noneβ€”β€”game-theorycoordination+14 articlesView β†’
πŸ”— WebSANS Security Awarenessβ€”Noneβ€”β€”cybersecuritysocial-engineering+21 articleView β†’
πŸ”— WebSB 25B-004 signedβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperScalable agent alignment via reward modelingarXivβ˜…β˜…β˜…β˜†β˜†SummaryJan Leike, David Krueger, Tom Everitt, Miljan Martic, Vishal Maini, Shane Legg2018-11-19alignmentcapabilities+32 articlesView β†’
πŸ”— WebScalable Human Oversight for Aligned LLMsβ€”Noneβ€”β€”alignmentllm1 articleView β†’
πŸ”— WebScalable Oversightβ€”Noneβ€”β€”traininghuman-feedback+32 articlesView β†’
✏️ Blog PostScalable Oversight and Weak-to-Strong GeneralizationAlignment Forumβ˜…β˜…β˜…β˜†β˜†MetadataAnsh Radhakrishnan, Buck, ryan_greenblatt, Fabien Roger2023-12-16β€”β€”View β†’
✏️ Blog PostScalable Oversight ApproachesAlignment Forumβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperScalable Oversight for Superhuman AI via Recursive Self-CritiquingarXivβ˜…β˜…β˜…β˜†β˜†SummaryXueru Wen, Jie Lou, Xinyu Lu, Junjie Yang, Yanjiang Liu, Yaojie Lu, Debing Zhang, Xing Yu2025-02-07alignmentcapabilities+32 articlesView β†’
✏️ Blog PostScalable Oversight in AI: Beyond Human SupervisionMediumβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”debaterecursive-reward-modeling+11 articleView β†’
πŸ“„ PaperScalable Oversight via Partitioned Human SupervisionarXivβ˜…β˜…β˜…β˜†β˜†SummaryRen Yin, Takashi Ishida, Masashi Sugiyama2025-10-26capabilitiestraining+31 articleView β†’
πŸ“„ Paperscalable oversight via recursive self-critiquingarXivβ˜…β˜…β˜…β˜†β˜†SummaryXueru Wen, Jie Lou, Xinyu Lu, Junjie Yang, Yanjiang Liu, Yaojie Lu, Debing Zhang, Xing Yu2025-02-07alignmentcapabilities+21 articleView β†’
πŸ“„ PaperScalable watermarking for identifying large language model outputsNatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”llmdeepfakes+21 articleView β†’
πŸ”— WebScale Adversarial Robustness Leaderboard

A comprehensive evaluation framework testing large language models' resistance to adversarial prompts across multiple harm categories. Ranks models based on their ability to avoid generating harmful responses.

β€”Fullβ€”β€”evaluationllmβ€”View β†’
πŸ”— WebScale AI Adversarial Robustnessβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebScale AI's SWE-Bench Proβ€”Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ“„ PaperScale limitationsarXivβ˜…β˜…β˜…β˜†β˜†SummaryKevin Wang, Alexandre Variengien, Arthur Conmy, Buck Shlegeris, Jacob Steinhardt2022-11-01interpretabilityevaluation+11 articleView β†’
πŸ“„ PaperScaling Laws For Scalable OversightarXivβ˜…β˜…β˜…β˜†β˜†SummaryJoshua Engels, David D. Baek, Subhash Kantamneni, Max Tegmark2025-04-25capabilitiesagi1 articleView β†’
πŸ“„ PaperScaling Laws For Scalable OversightarXivβ˜…β˜…β˜…β˜†β˜†SummaryJoshua Engels, David D. Baek, Subhash Kantamneni, Max Tegmark2025-04-25capabilitiesagi+31 articleView β†’
πŸ“„ PaperScaling Laws For Scalable OversightarXivβ˜…β˜…β˜…β˜†β˜†SummaryJoshua Engels, David D. Baek, Subhash Kantamneni, Max Tegmark2025-04-25capabilitiesagiβ€”View β†’
πŸ”— WebScaling MonosemanticityTransformer Circuits

The study demonstrates that sparse autoencoders can extract meaningful, abstract features from large language models, revealing complex internal representations across domains like programming, geography, and personal histories.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”interpretabilitycapabilities+36 articlesView β†’
πŸ”— WebScharre (2018) "Army of None"Amazonβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— WebSchelling Strategy of Conflictβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebScheming researchAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deceptionprioritization+21 articleView β†’
πŸ“„ PaperSchoenegger et al. (2024): AI ForecastingarXiv

SMORE is a resource-efficient domain adaptation algorithm using hyperdimensional computing to dynamically customize test-time models. It achieves higher accuracy and faster performance compared to existing deep learning approaches.

β˜…β˜…β˜…β˜†β˜†FullWang, Junyao, Faruque, Mohammad Abdullah Al2024capabilitiesforecasting+21 articleView β†’
πŸ”— Webscholar.google.comGoogle Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”geminialphafold+11 articleView β†’
πŸ”— WebSchwartz Reisman Instituteβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperScienceScienceβ˜…β˜…β˜…β˜…β˜…MetadataB. Sparrow, Jenny J W Liu, D. Wegner2011-08-05human-agencyautomation+11 articleView β†’
πŸ“„ PaperScience AdvancesScienceβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”β€”View β†’
πŸ“„ PaperScience Magazine EditorialScienceβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”market-concentrationgovernance+12 articlesView β†’
πŸ“„ PaperScience: Fake papersScienceβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”β€”View β†’
πŸ”— WebScienceDirectScienceDirectβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”power-seekingself-preservation+13 articlesView β†’
πŸ”— WebScientific American, 2024β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebScientific Panelβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSEC whistleblower complaintβ€”Noneβ€”β€”frontier-labssafety-culture+11 articleView β†’
πŸ”— WebSecBenchGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebSecond Talent - Chinese AI Investment Statistics 2025

China invested $125 billion in AI in 2025, representing 38% of global investment, with significant government backing and concentration in autonomous vehicles, computer vision, and strategic technologies.

β€”Fullβ€”β€”compute1 articleView β†’
πŸ›οΈ GovernmentSection 1066 of the FY2025 NDAAUS Congressβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebSecureBio organizationβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebSecurity challenges by AI-assisted protein designβ€”Noneβ€”β€”cybersecuritybiosecurity+21 articleView β†’
πŸ“„ PaperSecurity researcharXivβ˜…β˜…β˜…β˜†β˜†SummaryBrendan Murphy, Dillon Bowen, Shahrad Mohammadzadeh, Tom Tseng, Julius Broomfield, Adam Gleave, Kellin Pelrine2025-07-15governancecapabilities+31 articleView β†’
πŸ“„ PaperSelf-correction researcharXivβ˜…β˜…β˜…β˜†β˜†SummaryJΓ©rΓ©my Scheurer, Jon Ander Campos, Tomasz Korbak, Jun Shern Chan, Angelica Chen, Kyunghyun Cho, Ethan Perez2023-03-28capabilitiesevaluation+31 articleView β†’
πŸ”— WebSelf-modeling is instrumentally usefulAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”probabilitydecomposition+11 articleView β†’
πŸ”— WebSeligman (1972)β€”Noneβ€”β€”information-overloadmedia-literacy+11 articleView β†’
πŸ”— WebSeligman: Learned Helplessness Original Researchβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebSemantic ScholarSemantic Scholar

Semantic Scholar is a free, AI-powered research platform that enables comprehensive scientific literature search and discovery. The tool aims to make academic research more accessible and contextual.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebSemantic Scholar'sSemantic Scholarβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebSEMIβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSEMI fab capacity reportβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperSemi-Synthetic Transformers for Evaluating Mechanistic Interpretability TechniquesarXivβ˜…β˜…β˜…β˜†β˜†SummaryRohan Gupta, IvΓ‘n Arcuschin, Thomas Kwa, AdriΓ  Garriga-Alonso2024-07-19interpretabilitycapabilities+31 articleView β†’
πŸ”— WebSemiAnalysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSemiAnalysis Huawei production

SemiAnalysis examines Huawei's AI chip production capabilities, highlighting challenges from export controls and memory bottlenecks. The analysis reveals China's strategic efforts to develop domestic semiconductor manufacturing.

β€”Fullβ€”β€”capabilitiescompute1 articleView β†’
πŸ›οΈ GovernmentSenate AI Insight Forumβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentSenate Bill 1047β€”Noneβ€”β€”regulationstate-policy+32 articlesView β†’
πŸ›οΈ GovernmentSenate Intelligence Committee Report

The Senate Intelligence Committee report details how the Internet Research Agency (IRA) used social media platforms to spread disinformation and divisive content targeting American voters during the 2016 election.

β€”Fullβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ›οΈ GovernmentSenate Judiciary Committee Chair Chuck Grassley introduced the AI Whistleblower Protection Actβ€”Noneβ€”β€”frontier-labssafety-culture+11 articleView β†’
πŸ›οΈ GovernmentSenate Testimony 2023β€”Noneβ€”β€”constitutional-airesponsible-scaling+11 articleView β†’
πŸ›οΈ GovernmentSenate testimony, 2023β€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentSenator Elizabeth Warren's proposed legislationβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ›οΈ GovernmentSenator Scott Wienerβ€”Noneβ€”β€”regulationstate-policy+11 articleView β†’
πŸ›οΈ GovernmentSenator Wiener: Bipartisan Vote, Senate Passes Landmark AI Safety Billβ€”Noneβ€”β€”safetyregulation+21 articleView β†’
πŸ”— WebSenseTimeReutersβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebSensity AI (Deepfake Detection Research)β€”Noneβ€”β€”mental-healthai-ethics+32 articlesView β†’
πŸ”— WebSensity AI: Deepfake analysisβ€”Noneβ€”β€”deepfakesdigital-evidence+35 articlesView β†’
πŸ›οΈ GovernmentSeoul AI Safety SummitUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyinternational+24 articlesView β†’
πŸ›οΈ GovernmentSeoul Declarationβ€”Noneβ€”β€”β€”3 articlesView β†’
πŸ›οΈ GovernmentSeoul Declaration for Safe, Innovative and Inclusive AIUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety3 articlesView β†’
πŸ”— WebSeoul Declaration on AI Safetyβ€”Noneβ€”β€”safetymonitoring+21 articleView β†’
πŸ›οΈ GovernmentSeoul Frontier AI CommitmentsUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”self-regulationindustry-commitments+35 articlesView β†’
πŸ›οΈ GovernmentSeoul Frontier AI Safety CommitmentsUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety1 articleView β†’
πŸ›οΈ GovernmentSeoul Statement of Intent toward International Cooperation on AI Safety ScienceUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety2 articlesView β†’
πŸ›οΈ GovernmentSeoul SummitUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorcompetition+11 articleView β†’
πŸ”— WebSequoia Capitalβ€”Noneβ€”β€”governancecoordination+11 articleView β†’
πŸ”— WebSesame Creditβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— Websetback reportedβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ“– Referenceseven former OpenAI employeesWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”ai-safetyconstitutional-ai+11 articleView β†’
πŸ“„ PaperSevilla et al.arXivβ˜…β˜…β˜…β˜†β˜†SummaryJaime Sevilla, Lennart Heim, Anson Ho, Tamay Besiroglu, Marius Hobbhahn, Pablo Villalobos2022-02-11capabilitiestraining+31 articleView β†’
πŸ“„ PaperShavit et al. (2023)arXivβ˜…β˜…β˜…β˜†β˜†SummaryLewis Ho, Joslyn Barnhart, Robert Trager, Yoshua Bengio, Miles Brundage, Allison Carnegie, Rumman Chowdhury, Allan Dafoe, Gillian Hadfield, Margaret Levi, Duncan Snidal2023-07-10governancecapabilities+31 articleView β†’
πŸ“„ PaperShen, H., Knearem, T., Ghosh, R., et al. (2024). "Towards Bidirectional Human-AI Alignment: A Systematic Review."arXivβ˜…β˜…β˜…β˜†β˜†SummaryHua Shen, Tiffany Knearem, Reshmi Ghosh, Kenan Alkiek, Kundan Krishna, Yachuan Liu, Ziqiao Ma, Savvas Petridis, Yi-Hao Peng, Li Qiwei, Sushrita Rakshit, Chenglei Si, Yutong Xie, Jeffrey P. Bigham, Frank Bentley, Joyce Chai, Zachary Lipton, Qiaozhu Mei, Rada Mihalcea, Michael Terry, Diyi Yang, Meredith Ringel Morris, Paul Resnick, David Jurgens2024-06-13alignmentshutdown-problem+21 articleView β†’
πŸ“„ PaperShlegeris et al. (2024)arXivβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”monitoringcontainment+33 articlesView β†’
πŸ”— WebShorenstein Center: Platform Accountability

The Shorenstein Center examines media platforms, information distribution, and civic engagement across multiple domains. It focuses on understanding technological and policy impacts on news creation, consumption, and public discourse.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ“„ PaperShortcut Learning in Deep Neural NetworksarXivβ˜…β˜…β˜…β˜†β˜†SummaryRobert Geirhos, JΓΆrn-Henrik Jacobsen, Claudio Michaelis, Richard Zemel, Wieland Brendel, Matthias Bethge, Felix A. Wichmann2020-04-16capabilitiesevaluation+11 articleView β†’
πŸ”— WebShorter timelines than bio anchors suggestedβ€”Noneβ€”β€”effective-altruismai-safety-funding+11 articleView β†’
πŸ”— WebShoshana Zuboff (Harvard)β€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebShoshana Zuboff on Surveillance Capitalismβ€”Noneβ€”β€”β€”β€”View β†’
πŸ“„ PaperShutdown Resistance in Large Language ModelsarXivβ˜…β˜…β˜…β˜†β˜†SummaryJeremy Schlatter, Benjamin Weinstein-Raun, Jeffrey Ladish2025-09-13llmshutdown-problem+21 articleView β†’
πŸ“„ PaperShutdown Resistance in LLMsarXivβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”llm1 articleView β†’
πŸ”— WebShutdown-seeking AISpringer

The authors propose a novel AI safety approach of creating shutdown-seeking AIs with a final goal of being shut down. This strategy aims to prevent dangerous AI behaviors by designing agents that will self-terminate if they develop harmful capabilities.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiessafety+32 articlesView β†’
πŸ”— WebSidley: New U.S. Export Controls on AI

The Bureau of Industry and Security (BIS) published updated export regulations targeting advanced computing items and AI model weights, significantly expanding control mechanisms for international technology transfers.

β€”Fullβ€”β€”governancecybersecurity1 articleView β†’
πŸ”— WebSidley: Rising AI Enforcement Insights

State and federal authorities are increasing scrutiny of AI technologies, targeting deceptive marketing claims and potential biases in AI products across various sectors.

β€”Fullβ€”β€”deceptionβ€”View β†’
πŸ”— WebSightsource Manufacturing ROI

The document explores how AI technologies can transform manufacturing operations by addressing quality control, predictive maintenance, and decision-making inefficiencies. It provides a comprehensive overview of AI implementation strategies with detailed ROI and implementation considerations.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebSignal Protocolβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ›οΈ GovernmentSigned Bill Text (PDF)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ Governmentsigned by Governor Bill Lee on March 21, 2024β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSigned into law by Governor Jared Polisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Websigned TRAIGA into lawβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSiliconANGLE, 2023β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSimilar proposals in the EUβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebSimilarweb Traffic Dataβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ”— WebSimpleQAOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSingapore Consensus

I apologize, but I cannot generate a meaningful summary for this source. The provided content appears to be an incomplete or corrupted HTML fragment, specifically containing only a Google Tag Manager iframe, which does not provide any substantive text or information about the "Singapore Consensus". To properly analyze a source document, I would need: - Full readable text content - Clear context about the document's subject - Actual substantive information about the topic Would you be able to provide the complete text of the "Singapore Consensus" document? Without the actual content, I cannot construct a valid summary following the requested JSON format.

β€”Summaryβ€”β€”β€”β€”View β†’
πŸ”— WebSingapore Statementβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ”— WebSIPRI researchβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ Papersituational awarenessarXivβ˜…β˜…β˜…β˜†β˜†SummaryLukas Berglund, Asa Cooper Stickland, Mikita Balesni, Max Kaufmann, Meg Tong, Tomasz Korbak, Daniel Kokotajlo, Owain Evans2023-09-01alignmentcapabilities+33 articlesView β†’
πŸ“„ PaperSkalse et al. (2022)arXivβ˜…β˜…β˜…β˜†β˜†SummaryJoar Skalse, Nikolaus H. R. Howe, Dmitrii Krasheninnikov, David Krueger2022-09-27specification-gaminggoodharts-law+11 articleView β†’
πŸ”— WebSlatestar Codex: Meditations on Molochβ€”Noneβ€”β€”game-theorycoordination+11 articleView β†’
πŸ”— WebSleeper AgentsAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSleeper Agentsβ€”Noneβ€”β€”ai-safetyconstitutional-ai+11 articleView β†’
πŸ“„ PaperSleeper AgentsarXiv

A study exploring deceptive behavior in AI models by creating backdoors that trigger different responses based on context. The research demonstrates significant challenges in removing such deceptive strategies using standard safety training methods.

β˜…β˜…β˜…β˜†β˜†FullHubinger, Evan, Denison, Carson, Mu, Jesse, Lambert, Mike, Tong, Meg, MacDiarmid, Monte, Lanham, Tamera, Ziegler, Daniel M., Maxwell, Tim, Cheng, Newton, Jermyn, Adam, Askell, Amanda, Radhakrishnan, Ansh, Anil, Cem, Duvenaud, David, Ganguli, Deep, Barez, Fazl, Clark, Jack, Ndousse, Kamal, Sachan, Kshitij, Sellitto, Michael, Sharma, Mrinank, DasSarma, Nova, Grosse, Roger, Kravec, Shauna, Bai, Yuntao, Witten, Zachary, Favaro, Marina, Brauner, Jan, Karnofsky, Holden, Christiano, Paul, Bowman, Samuel R., Graham, Logan, Kaplan, Jared, Mindermann, SΓΆren, Greenblatt, Ryan, Shlegeris, Buck, Schiefer, Nicholas, Perez, Ethan2024safetydeception+315 articlesView β†’
πŸ”— WebSMM ASML lead times

I apologize, but the provided content does not appear to be a substantive source document about AI safety or anything meaningful. The text seems to be a jumbled list of market and industry categories, website navigation links, and a legal notice. Without a coherent source document, I cannot complete the requested summary in the specified JSON format. To proceed, I would need: 1. A clear, readable source document 2. A substantive text discussing a specific topic or research finding 3. Ideally, a document related to AI safety, machine learning, or technological risk Would you like to provide an alternative source document for analysis?

β€”Summaryβ€”β€”safety1 articleView β†’
πŸ”— WebSoaresMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”causal-modelcorrigibility+11 articleView β†’
πŸ”— WebSocial Credit Systemβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebSolace Global - Escalation of the US-China AI Arms Race in 2025

The US and China are competing for AI technological supremacy, with export controls and geopolitical tensions significantly impacting AI development and strategic capabilities.

β€”Fullβ€”β€”capabilities1 articleView β†’
✏️ Blog PostSome experts like Eliezer YudkowskyLessWrongβ˜…β˜…β˜…β˜†β˜†MetadataEliezer Yudkowsky2022-04-02x-riskirreversibility+11 articleView β†’
✏️ Blog Postsome researchers noteAlignment Forumβ˜…β˜…β˜…β˜†β˜†Metadatascasper2024-05-21β€”1 articleView β†’
πŸ”— Websome researchers noteβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSora qualityOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebSORMβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebSouth China Morning Post - China's AI capital spending 2025

A Bank of America report forecasts China's AI capital expenditure to grow 48% in 2025, with total spending between US$84-98 billion. Government and major tech companies are driving substantial investments in AI technology.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebSovereign Internet Lawβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ“– ReferenceSoviet biological weapons programWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebSPARβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSPAR - Research Program for AI Risks

SPAR is a research program that pairs mentees with experienced professionals to work on AI safety, policy, and related research projects. The program offers structured research experience, mentorship, and potential publication opportunities.

β€”Fullβ€”β€”governancesafety+31 articleView β†’
πŸ”— WebSpark analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSparse AutoencodersAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperSparse AutoencodersarXivβ˜…β˜…β˜…β˜†β˜†SummaryLeonard Bereska, Efstratios Gavves2024-04-22alignmentinterpretability+34 articlesView β†’
πŸ“„ PaperSparse Autoencoders Find Highly Interpretable Features in Language ModelsarXivβ˜…β˜…β˜…β˜†β˜†SummaryHoagy Cunningham, Aidan Ewart, Logan Riggs, Robert Huben, Lee Sharkey2023-09-15interpretabilityeconomic+31 articleView β†’
πŸ”— WebSparse Autoencoders for Interpretability in Reinforcement Learning Modelsβ€”Noneβ€”β€”interpretabilitysparse-autoencoders+21 articleView β†’
πŸ”— WebSparse autoencoders uncover biologically interpretable features in protein language model representationsPNASβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”interpretabilitybiosecurity+31 articleView β†’
πŸ”— Webspec.c2pa.orgβ€”Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ”— WebSpecification gaming examplesGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”geminialphafold+11 articleView β†’
πŸ”— Webspecification gaming examples databaseβ€”Noneβ€”β€”taxonomyreward-modeling+32 articlesView β†’
πŸ”— WebSpecification Gaming: The Flip Side of AI IngenuityGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”taxonomyreward-modeling+33 articlesView β†’
πŸ”— WebSpherical Insights: Top 10 AI Spending Countries 2025

A comprehensive analysis of the top 10 countries investing in AI technology in 2025, revealing significant national commitments to AI development and innovation.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebSQ Magazine

In 2025, AI is driving significant job creation globally, generating 97 million new roles while displacing 85 million jobs. The net effect is a positive transformation of the workforce across industries and skill levels.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebSquiggle

Squiggle is a programming library for working with probability distributions in JavaScript/Rescript. It provides efficient tools for probabilistic calculations with minimal computational overhead.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebSt. Louis Fed Analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebStability AIβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebStable Diffusion XLβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebStack Overflow Developer Surveyβ€”Noneβ€”β€”networksrisk-interactions+11 articleView β†’
πŸ”— WebStaged deploymentAnthropicβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”defensesecurity+11 articleView β†’
πŸ“„ PaperStandardised Schema for AI Incident DatabasesarXivβ˜…β˜…β˜…β˜†β˜†SummaryAvinash Agarwal, Manisha J. Nene2025-01-28governancesafety1 articleView β†’
πŸ”— WebStanford AI Index

The annual AI Index report provides comprehensive insights into AI trends, including increased regulations, generative AI investment, and model training complexities. It covers technical, economic, and societal dimensions of AI development.

β€”Fullβ€”β€”governancetraining+11 articleView β†’
πŸ”— WebStanford AI Index 2025Stanford HAI

The 2025 AI Index Report documents massive growth in global AI private investment, with the U.S. leading in funding and organizational AI adoption reaching 78%. The report highlights transformative impacts across business functions and technological domains.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebStanford AI Index 2025Stanford HAI

The 2025 AI Index Report from Stanford HAI offers a detailed analysis of AI's technological, economic, and social developments. It highlights key trends in performance, investment, global leadership, and responsible AI adoption.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitieseconomic3 articlesView β†’
πŸ”— WebStanford AI Index Report 2025β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebStanford AI Safetyβ€”Noneβ€”β€”safetycorrigibility+21 articleView β†’
πŸ”— WebStanford Deliberative Democracy Lab

The lab focuses on deliberative democracy techniques to engage citizens in meaningful discussions about emerging technologies and social issues, with a particular emphasis on AI governance and public participation.

β€”Fullβ€”β€”governancedemocratic-innovation+11 articleView β†’
πŸ”— WebStanford Deliberative Polling Timelineβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebStanford Education Research (2023)β€”Noneβ€”β€”information-overloadmedia-literacy+11 articleView β†’
πŸ”— WebStanford FSI - Getting Ahead of Digital Repression

A comprehensive analysis of how authoritarian states, particularly China, are developing and exporting digital technologies for social control and repression. The report examines emerging technologies' potential for undermining democratic freedoms.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebStanford FSI - Government Venture Capital and AI Development in China

China's government VC funds have invested heavily in AI, distributing capital more evenly across regions than private VCs. These investments often precede and signal opportunities for private venture capital.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebStanford FSI: Digital Repression Research

The Freeman Spogli Institute (FSI) at Stanford is a hub for nonpartisan international research, teaching, and policy impact across various global domains.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ”— WebStanford FSI: Regulating Under Uncertaintyβ€”Noneβ€”β€”governance1 articleView β†’
πŸ”— WebStanford HAIStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”regulationgpai+11 articleView β†’
πŸ”— WebStanford HAIStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebStanford HAIβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebStanford HAI (2024)Stanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”social-engineeringmanipulation+11 articleView β†’
πŸ”— WebStanford HAI AI IndexStanford HAI

A comprehensive global survey examining public perceptions of AI across 26 nations, tracking changes in attitudes towards AI's benefits, risks, and potential impacts on society and work.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebStanford HAI frameworkβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebStanford HAI researchStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cascadesrisk-pathways+11 articleView β†’
πŸ”— WebStanford HAI ResearchStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”networksrisk-interactions+11 articleView β†’
πŸ”— WebStanford HAI studyStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+11 articleView β†’
πŸ”— WebStanford HAI: 2025 AI Index Report - Policy and GovernanceStanford HAI

The 2025 AI Index Report highlights significant growth in AI-related legislation, government investments, and international safety collaboration across multiple countries.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”governancesafetyβ€”View β†’
πŸ”— WebStanford HAI: AI Companions and Mental HealthStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”timelineautomation+316 articlesView β†’
πŸ”— WebStanford HAI: The Disinformation MachineStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”β€”View β†’
πŸ”— WebStanford HAI's implementation trackerStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”compute-thresholdsgovernance+12 articlesView β†’
πŸ”— WebStanford Healthcare's radiology AIβ€”Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ”— WebStanford History Education Groupβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebStanford Internet Observatory

Stanford's Cyber Policy Center conducts interdisciplinary research on technology's impact on governance, democracy, and public policy. The center hosts seminars and produces research across various digital policy domains.

β€”Fullβ€”β€”governancecybersecurity+33 articlesView β†’
πŸ”— WebStanford Internet Observatory

Stanford's Cyber Policy Center is an interdisciplinary research center studying technology's impact on governance, democracy, and public policy. It hosts seminars, conducts research, and explores emerging digital challenges.

β€”Fullβ€”β€”governancecybersecurityβ€”View β†’
πŸ”— WebStanford Internet Observatoryβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebStanford Persuasive Technology Labβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebStanford RegLab: AI Regulationβ€”Noneβ€”β€”governancemental-health+21 articleView β†’
πŸ”— WebStanford researchStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”scalingcapability-evaluation+11 articleView β†’
πŸ”— WebStanford researchStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ”— WebStanford studyStanford HAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”timelinecapability+11 articleView β†’
πŸ“„ PaperStanford: Detecting AI-generated text unreliablearXiv

This Stanford study explores the vulnerabilities of AI text detection techniques by developing recursive paraphrasing attacks that significantly reduce detection accuracy across multiple detection methods with minimal text quality degradation.

β˜…β˜…β˜…β˜†β˜†FullSadasivan, Vinu Sankar, Kumar, Aounon, Balasubramanian, Sriram, Wang, Wenxiao, Feizi, Soheil2025cybersecurityepistemic+32 articlesView β†’
πŸ”— WebStanford's Alpaca projectβ€”Noneβ€”β€”open-sourcegovernance+11 articleView β†’
πŸ”— WebStanford's Platform Governance Archiveβ€”Noneβ€”β€”governanceβ€”View β†’
πŸ”— WebStanislav Petrov's decisionβ€”Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— WebStarkWareβ€”Noneβ€”β€”game-theorygovernance+11 articleView β†’
πŸ”— WebStarling Bank researchβ€”Noneβ€”β€”disinformationdeepfakes+11 articleView β†’
πŸ”— WebStatcheckβ€”Noneβ€”β€”scientific-integritypaper-mills+11 articleView β†’
πŸ›οΈ GovernmentState attorney general investigationβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebState of AI Report 2025

The annual State of AI Report examines key developments in AI research, industry, politics, and safety for 2025, featuring insights from a large-scale practitioner survey.

β€”Fullβ€”β€”safetyrisk-factor+33 articlesView β†’
πŸ”— Webstating in his signing statementβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebStatistaβ€”Noneβ€”β€”disinformationastroturfing+11 articleView β†’
πŸ”— WebStatista market dataβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSteve Omohundro (2008)β€”Noneβ€”β€”β€”1 articleView β†’
πŸ“– ReferenceSteve Omohundro's seminal work on "basic AI drives"Wikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”corrigibilityshutdown-problem+13 articlesView β†’
πŸ”— WebStop Killer Robots Campaign Videosβ€”Noneβ€”β€”escalationconflict+32 articlesView β†’
πŸ”— WebStratechery Analysisβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— Webstrategic game-playing systemsGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”formal-analysispower-seeking+11 articleView β†’
πŸ”— Webstrategic insights from simulation gaming of AI race dynamicsScienceDirectβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperStrategic Insights from Simulation Gaming of AI Race DynamicsarXivβ˜…β˜…β˜…β˜†β˜†SummaryRoss Gruetzemacher, Shahar Avin, James Fox, Alexander K Saeri2024-10-04governancecapabilities+31 articleView β†’
πŸ”— WebStrengthening the Biological Weapons Conventionβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— Webstripped by a near-unanimous 99-1 Senate voteβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebStuart Russellβ€”Noneβ€”β€”frameworkinstrumental-goals+34 articlesView β†’
πŸ”— WebStuart Russellβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebStuart Russell, UC Berkeleyβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebStudies indicateSAGE Journalsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebStudies of nuclear arms controlβ€”Noneβ€”β€”game-theoryinternational-coordination+12 articlesView β†’
πŸ”— Webstudies on risk perceptions across the Global North and SouthSAGE Journalsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— Webstudy by Berkeley researchersβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— Webstudy showed poor performanceβ€”Noneβ€”β€”capabilitieshuman-agency+21 articleView β†’
πŸ”— Websubsequent researchβ€”Noneβ€”β€”ai-biasalgorithmic-accountability+11 articleView β†’
πŸ“„ PaperSunstein, Constitutional Political Economy (2018)Springerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebSuperalignment teamOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”research-agendasalignment+34 articlesView β†’
✏️ Blog PostSuperforecasting the Premises in 'Is Power-Seeking AI an Existential Risk?'β€”SummaryJoseph Carlsmith2023-10-18forecastingx-risk+11 articleView β†’
πŸ“– ReferenceSuperintelligenceWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”agix-risk+23 articlesView β†’
πŸ”— WebSuperintelligence: Paths, Dangers, StrategiesAmazonβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”agischeming+21 articleView β†’
πŸ“– ReferenceSurvey of AI researchersWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”power-seekingself-preservation+13 articlesView β†’
πŸ”— WebSurvival and Flourishing Fund

SFF is a virtual fund that organizes grant recommendations and philanthropic giving, primarily supporting organizations working on existential risk and AI safety. They use a unique S-Process and have distributed over $152 million in grants since 2019.

β€”Fullβ€”β€”safetyx-risk+31 articleView β†’
πŸ”— WebSusser et al. 2019SSRNβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebSWE-bench Official Leaderboards

SWE-bench provides a multi-variant evaluation platform for assessing AI models' performance in software engineering tasks. It offers different datasets and metrics to comprehensively test AI coding agents.

β€”Fullβ€”β€”capabilitiesevaluation+35 articlesView β†’
πŸ”— WebSWE-bench Pro Leaderboard - Scale AI

SWE-Bench Pro provides a comprehensive evaluation of AI agents' software engineering skills by sourcing tasks from public and private repositories. The benchmark addresses key limitations in existing benchmarks by focusing on realistic, challenging problem-solving scenarios.

β€”Fullβ€”β€”capabilitiesevaluationβ€”View β†’
πŸ”— WebSWE-bench Verified - OpenAIOpenAI

OpenAI collaborated with software developers to improve the SWE-bench benchmark by identifying and filtering out problematic test samples. The resulting SWE-bench Verified provides a more reliable evaluation of AI models' software engineering skills.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiesevaluation+32 articlesView β†’
πŸ“„ PaperSWE-bench: Can Language Models Resolve Real-World GitHub Issues?arXivβ˜…β˜…β˜…β˜†β˜†SummaryCarlos E. Jimenez, John Yang, Alexander Wettig, Shunyu Yao, Kexin Pei, Ofir Press, Karthik Narasimhan2023-10-10capabilitiestraining+31 articleView β†’
πŸ”— WebSwiss foundationβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebSycophancy in Generative-AI Chatbotsβ€”Noneβ€”β€”alignmenttruthfulness+11 articleView β†’
πŸ“„ Papersycophancy in LLMsarXivβ˜…β˜…β˜…β˜†β˜†SummaryLars Malmqvist2024-11-22alignmentcapabilities+31 articleView β†’
✏️ Blog Postsycophancy is more likely than schemingLessWrongβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”deceptionβ€”View β†’
πŸ”— WebSynergy Researchβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebSynthesiaβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebSynthID - Google DeepMindGoogle DeepMindβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deepfakesdigital-evidence+32 articlesView β†’
πŸ“„ PaperSynthID-Image: Image watermarking at internet scalearXivβ˜…β˜…β˜…β˜†β˜†SummarySven Gowal, Rudy Bunel, Florian Stimberg, David Stutz, Guillermo Ortiz-Jimenez, Christina Kouridi, Mel Vecerik, Jamie Hayes, Sylvestre-Alvise Rebuffi, Paul Bernard, Chris Gamble, MiklΓ³s Z. HorvΓ‘th, Fabian Kaczmarczyck, Alex Kaskasoli, Aleksandar Petrov, Ilia Shumailov, Meghana Thotakuri, Olivia Wiles, Jessica Yung, Zahra Ahmed, Victor Martin, Simon Rosen, Christopher Savčak, Armin Senoner, Nidhi Vyas, Pushmeet Kohli2025-10-10capabilitiesevaluation+31 articleView β†’
πŸ”— WebSystem-level safety approachMeta AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyinternational+21 articleView β†’
πŸ“„ Papersystematic evaluation of medical vision-language modelsarXivβ˜…β˜…β˜…β˜†β˜†SummaryZikun Guo, Jingwei Lv, Xinyue Xu, Shu Yang, Jun Wen, Di Wang, Lijie Hu2025-09-26evaluationllm1 articleView β†’
πŸ”— Websystematic review of healthcare ML (2025)ScienceDirectβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”robustnessgeneralization+11 articleView β†’
πŸ”— WebSystematic review on neural architecture searchSpringerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”intelligence-explosionrecursive-self-improvement+11 articleView β†’
πŸ“„ PaperSystemic Risk in AI DevelopmentarXivβ˜…β˜…β˜…β˜†β˜†SummaryNathakhun Wiroonsri, Onthada Preedasawakul2023-08-28capabilitiesevaluation+31 articleView β†’
πŸ›οΈ GovernmentSystemic Safety GrantsUK AI Safety Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebTaiwan Alignment Assemblyβ€”Noneβ€”β€”alignmentdemocratic-innovation+21 articleView β†’
πŸ”— WebTaiwan's digital democracy initiativesβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ“„ PaperTake Goodhart Seriously: Principled Limit on General-Purpose AI OptimizationarXivβ˜…β˜…β˜…β˜†β˜†SummaryAntoine Maier, Aude Maier, Tom David2025-10-03taxonomyreward-modeling+11 articleView β†’
πŸ”— WebTech Policy Press

An online publication covering technology policy issues, featuring analysis, perspectives, and discussions on digital governance, AI, online safety, and related policy challenges.

β€”Fullβ€”β€”governancesafety+31 articleView β†’
πŸ”— Webtech sector data from 2024-25β€”Noneβ€”β€”labor-marketsautomation+11 articleView β†’
πŸ”— WebTech Startups - AI investments make up 33% of total U.S. venture capital funding in 2024

AI investments are dominating venture capital, rising from 14% in 2020 to 33% in 2024, with major investments concentrated in foundational AI model development.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebTechCongressβ€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebTechCrunchTechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”3 articlesView β†’
πŸ”— WebTechCrunch CoverageTechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”cognitive-emulationcoem+11 articleView β†’
πŸ”— WebTechCrunch ReportsTechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”cognitive-emulationcoem+32 articlesView β†’
πŸ”— WebTechCrunch, 2024TechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebTechCrunch: California's legislature just passed AI bill SB 1047TechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”regulationstate-policy+11 articleView β†’
πŸ”— WebTechCrunch: ChatGPT 2024 TimelineTechCrunch

OpenAI's ChatGPT experienced significant growth and product evolution in 2024, including partnerships with Apple, enterprise expansions, and new AI model releases like GPT-4o.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”open-sourcellmβ€”View β†’
πŸ”— WebTechInsights Q1 2024

TechInsights reports on the explosive growth of the data-center AI chip market in 2023, highlighting NVIDIA's market leadership and revenue surge.

β€”Fullβ€”β€”compute1 articleView β†’
πŸ”— Webtechnical infrastructure as a hidden terrain of disinformationβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— Webtechnical paperβ€”Noneβ€”β€”intelligence-explosionrecursive-self-improvement+11 articleView β†’
πŸ”— WebTechnical Performance - 2025 AI Index ReportStanford HAI

The 2025 AI Index Report highlights dramatic improvements in AI model performance, including faster benchmark mastery, convergence of model capabilities, and emerging reasoning paradigms.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiesevaluation+31 articleView β†’
πŸ”— WebTechnical specificationβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebTechnology to Secure the AI Chip Supply ChainCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”compute2 articlesView β†’
πŸ”— WebTechPolicy.Pressβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebTechPolicy.Press analysisβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebTechPolicy.Press analysisβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebTED Talksβ€”Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ“„ PaperTEE for ML (Chen et al.)arXivβ˜…β˜…β˜…β˜†β˜†SummaryFatemeh Hashemniya, BenoΓ―t Caillaud, Erik Frisk, Mattias Krysander, Mathias Malandain2023-12-21game-theorygovernance+11 articleView β†’
πŸ”— WebTenableβ€”Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebTest Scores AI vs Humans - Our World in DataOur World in Data

A dataset tracking AI performance across various domains like language understanding, image recognition, and problem-solving. Provides a comparative framework for evaluating AI capabilities relative to human benchmarks.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”capabilitiesevaluationβ€”View β†’
πŸ›οΈ GovernmentTesting Risks of AI for National Security (TRAINS) Taskforceβ€”Noneβ€”β€”evaluationcybersecurity1 articleView β†’
πŸ”— WebTetlock research

Philip Tetlock's research on Superforecasting reveals a group of experts who consistently outperform traditional forecasting methods by applying rigorous analytical techniques and probabilistic thinking.

β€”Fullβ€”β€”forecastingprediction-markets+32 articlesView β†’
πŸ”— WebTetlock: Superforecasting

Philip Tetlock's research on superforecasting demonstrates how careful probabilistic thinking and systematic approaches can significantly enhance forecasting accuracy in uncertain domains like epidemiology.

β€”Fullβ€”β€”forecastingprediction-markets+11 articleView β†’
πŸ›οΈ GovernmentTexas's SB 751β€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebThe AI and Biological Weapons ThreatRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”biosecuritygame-theory+34 articlesView β†’
πŸ”— WebThe AI Race Through a Geopolitical Lensβ€”Noneβ€”β€”game-theorycoordination+11 articleView β†’
πŸ”— WebThe AI Safety Institute International Network: Next StepsCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety6 articlesView β†’
πŸ”— WebThe AI Seoul SummitCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalgovernance+11 articleView β†’
πŸ›οΈ GovernmentThe Alignment Projectβ€”Noneβ€”β€”alignment1 articleView β†’
πŸ”— WebThe Associated Pressβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebThe Atlantic Council's trackingAtlantic Councilβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebThe Atlantic: "The AI That Agrees With Everything"β€”MetadataKatherine J. Wu2023β€”β€”View β†’
πŸ”— WebThe Atlantic: "The Doom Loop of Distrust"β€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebThe Atlantic: "The Epistemic Crisis"β€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebThe Atlantic: What Astroturfing Looks Like

An analysis of the 1918 influenza pandemic highlights strategies for managing public health during disease outbreaks, drawing parallels with modern pandemic responses.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebThe Authoritarian Risks of AI Surveillanceβ€”Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ”— WebThe Carnegie Endowment for International PeaceCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebThe Center for Strategic and International StudiesCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebThe Cyber-Biosecurity Nexusβ€”Noneβ€”β€”biosecuritycybersecurity+21 articleView β†’
πŸ”— WebThe Dangers of the Global Spread of China's Digital AuthoritarianismCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskgovernance+12 articlesView β†’
πŸ”— WebThe Diplomatβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebThe Economist: "Declining Trust"β€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebThe Economist: Tech Monopoliesβ€”Noneβ€”β€”β€”β€”View β†’
πŸ“„ PaperThe Energy Loss Phenomenon in RLHFarXiv

The study reveals an energy loss phenomenon during RLHF that correlates with reward hacking. The authors propose an Energy loss-aware PPO (EPPO) algorithm to mitigate this issue.

β˜…β˜…β˜…β˜†β˜†FullMiao, Yuchun, Zhang, Sen, Ding, Liang, Zhang, Yuqi, Zhang, Lefei, Tao, Dacheng2025trainingcybersecurityβ€”View β†’
πŸ›οΈ GovernmentThe FBI's 2024 Internet Crime Reportβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ›οΈ GovernmentThe Federal Communications Commission's investigationβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ›οΈ GovernmentThe Federal Election Commissionβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebThe Fight to Protect AI Whistleblowers

The provided text appears to be a collection of labor law and union-related news articles with no coherent focus on AI whistleblowers.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebThe Gap Between Open and Closed AI Models Might Be ShrinkingTIME

Epoch AI research reveals that open AI models are approximately one year behind closed models in capabilities, with the gap potentially shrinking as open models advance.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”capabilitiesopen-source1 articleView β†’
πŸ”— WebThe Global Landscape of AI Safety Institutesβ€”Noneβ€”β€”safetyfield-building+21 articleView β†’
πŸ”— WebThe Governance of AIFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governance1 articleView β†’
πŸ”— WebThe Guardian: Research fraudβ€”Noneβ€”β€”β€”β€”View β†’
πŸ›οΈ GovernmentThe History of Anthrax Weaponization in the Soviet Unionβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebThe Informationβ€”Noneβ€”β€”gpt-4chatgpt+12 articlesView β†’
πŸ”— WebThe Long, Contentious Battle to Regulate Gain-of-Function Workβ€”Noneβ€”β€”governancebiosecurity+21 articleView β†’
πŸ“„ PaperThe Malicious Use of AI - Future of Humanity InstitutearXivβ˜…β˜…β˜…β˜†β˜†SummaryMiles Brundage, Shahar Avin, Jack Clark, Helen Toner, Peter Eckersley, Ben Garfinkel, Allan Dafoe, Paul Scharre, Thomas Zeitzoff, Bobby Filar, Hyrum Anderson, Heather Roff, Gregory C. Allen, Jacob Steinhardt, Carrick Flynn, SeΓ‘n Γ“ hΓ‰igeartaigh, SJ Beard, Haydn Belfield, Sebastian Farquhar, Clare Lyle, Rebecca Crootof, Owain Evans, Michael Page, Joanna Bryson, Roman Yampolskiy, Dario Amodei2018-02-20cybersecurityopen-source+21 articleView β†’
πŸ“„ PaperThe MASK BenchmarkarXivβ˜…β˜…β˜…β˜†β˜†SummaryRichard Ren, Arunim Agarwal, Mantas Mazeika, Cristina Menghini, Robert Vacareanu, Brad Kenstler, Mick Yang, Isabelle Barrass, Alice Gatti, Xuwang Yin, Eduardo Trevino, Matias Geralnik, Adam Khoja, Dean Lee, Summer Yue, Dan Hendrycks2025-03-05capabilitiesdeception+2β€”View β†’
πŸ”— WebThe Neuron: Three Years of ChatGPT Retrospectiveβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebThe New York TimesThe New York Timesβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebThe News Literacy Project's comprehensive studyβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebThe OpenAI Safety Exodus: 25+ Senior Researchers Departed

Over 25 senior OpenAI researchers have departed, including key leadership in AI safety roles. The departures suggest a potential strategic realignment away from careful AI safety considerations.

β€”Fullβ€”β€”safety1 articleView β†’
πŸ”— WebThe Paradox of Augmentation: A Theoretical Model of AI-Induced Skill AtrophySSRNβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebThe Paris Summit: Au Revoir, global AI Safety?β€”Noneβ€”β€”safety1 articleView β†’
πŸ”— WebThe Partnership on AI's synthesis reportβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebThe Rise of DeepSeek: What the Headlines MissRAND Corporationβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebThe Rogue Replication Threat ModelMETRβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”interpretabilityscalable-oversight+11 articleView β†’
πŸ”— WebThe Safety Tax: How AI alignment reduces reasoning by up to 32%

Research reveals that AI safety techniques systematically degrade AI models' reasoning abilities. This 'Safety Tax' represents a significant challenge in developing responsible AI systems.

β€”Fullβ€”β€”alignmentsafetyβ€”View β†’
πŸ›οΈ GovernmentThe Securities and Exchange Commission's 2024 risk assessmentβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebThe Shallows: What the Internet Is Doing to Our Brainsβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebThe Social Dilemma (Netflix Documentary)β€”Noneβ€”β€”ai-ethicspersuasion+11 articleView β†’
πŸ”— WebThe Soufan Center: China-Russia Cooperation Analysis

The Soufan Center hosts an annual summit addressing terrorism and political violence, emphasizing the need to remain vigilant against evolving global threats.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebThe University of Washington's Center for an Informed Publicβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebThe upcoming ARC-AGI-2 benchmarkβ€”Noneβ€”β€”capabilitiesevaluation+31 articleView β†’
πŸ”— WebThe Verge AIβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebThe Verge: Courts and Deepfakesβ€”Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ”— WebTheoretical workβ€”Noneβ€”β€”causal-modelcorrigibility+32 articlesView β†’
πŸ”— WebThese reports create public accountabilityβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperTice et al. 2024arXivβ˜…β˜…β˜…β˜†β˜†SummaryCameron Tice, Philipp Alexander Kreer, Nathan Helm-Burger, Prithviraj Singh Shahani, Fedor Ryzhenkov, Fabien Roger, Clement Neo, Jacob Haimes, Felix HofstΓ€tter, Teun van der Weij2024-12-02governancecapabilities+31 articleView β†’
πŸ”— WebTiger Globalβ€”Noneβ€”β€”governancecoordination+11 articleView β†’
πŸ”— WebTikTok algorithm studyβ€”MetadataRob Barry, Georgia Wells, John West, Joanna Stern and Jason French2021ai-ethicspersuasion+11 articleView β†’
πŸ”— WebTikTok's algorithmβ€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebTIME - U.S. Military Spending on AI SurgesTIME

A Brookings Institution report reveals a massive increase in U.S. Department of Defense AI-related contracts, driven by technological advancements and geopolitical competition with China.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebTIME 2025TIMEβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”scalingcapability-evaluation+11 articleView β†’
πŸ”— WebTIME, 2024TIMEβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebTIME: China Is Taking AI Safety SeriouslyTIMEβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safetyregulation+22 articlesView β†’
πŸ”— WebTIME: How This Tool Could Decode AI's Inner MysteriesTIMEβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”mesa-optimizationβ€”View β†’
πŸ”— WebTimeline of AI model releases in 2024

VentureBeat created a detailed tracking of significant AI model releases in 2024, using data from the Artificial Intelligence Timeline project. The timeline covers both API and open weight models.

β€”Fullβ€”β€”open-source1 articleView β†’
πŸ”— Webtiming matters significantlyβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebTimnit Gebru et al.'s work

The Distributed AI Research Institute (DAIR) examines AI systems' societal impacts, emphasizing harm reduction and equitable technological futures. Their work centers on exposing systemic issues and developing alternative technological frameworks.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebTipping the ScalesCNASβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecurityinformation-warfare+11 articleView β†’
πŸ”— WebTNSR: Debunking the AI Arms Race Theoryβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebToby Ord's analysisFuture of Humanity Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebTom Davidson's compute-centric frameworkOpen Philanthropyβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”compute1 articleView β†’
πŸ”— WebTom's Hardwareβ€”Noneβ€”β€”compute1 articleView β†’
πŸ”— WebTom's Hardwareβ€”Noneβ€”β€”compute1 articleView β†’
πŸ”— WebTom's Hardware analysisβ€”Noneβ€”β€”compute1 articleView β†’
πŸ”— WebTom's Hardware ASML capacity

ASML, the world's leading lithography scanner manufacturer, is experiencing massive semiconductor equipment demand, with limited production capacity to meet current orders across market segments.

β€”Fullβ€”β€”compute1 articleView β†’
πŸ”— WebTom's Hardware China AI chip production

Chinese tech firms are ramping up domestic AI chip production to reduce dependence on foreign technologies. Their efforts face significant challenges in semiconductor fabrication and memory supply.

β€”Fullβ€”β€”compute1 articleView β†’
πŸ”— WebTom's Hardware H100 projections

Nvidia aims to significantly increase production of its H100 compute GPUs in 2024, driven by massive demand for AI and HPC applications. The company faces technical challenges in scaling production.

β€”Fullβ€”β€”capabilitiescompute1 articleView β†’
πŸ”— WebTony Blair Instituteβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webtool use and searchOpenAIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”power-seekingself-preservation+11 articleView β†’
πŸ“„ PaperToolEmuarXivβ˜…β˜…β˜…β˜†β˜†SummaryYangjun Ruan, Honghua Dong, Andrew Wang, Silviu Pitis, Yongchao Zhou, Jimmy Ba, Yann Dubois, Chris J. Maddison, Tatsunori Hashimoto2023-09-25capabilitiessafety+31 articleView β†’
πŸ”— WebTop 10 AI Research Papers of April 2025 (AryaXAI)

Ten landmark research papers examine the evolving landscape of AI, focusing on explainability, human-centered design, and responsible AI development across multiple domains.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebTop 50 AI Model Benchmarks 2025 - O-mega

This document provides an in-depth analysis of AI model benchmarks across multiple domains, highlighting how researchers evaluate AI capabilities through standardized tests and challenges.

β€”Fullβ€”β€”capabilitiesevaluationβ€”View β†’
πŸ”— WebTop AI Coding Models March 2025 - DEV Community

A comprehensive review of leading AI coding models, comparing their performance across benchmarks like HumanEval and SWE-bench. Models from Anthropic, OpenAI, Google, and others show remarkable progress in code generation and problem-solving.

β€”Fullβ€”β€”capabilitiesevaluationβ€”View β†’
πŸ”— WebTop Multimodal AI Models 2025

Multimodal AI models can process multiple types of data simultaneously, enabling more natural and contextually aware interactions across various applications and industries.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebTop OpenAI researcher resigns, saying company prioritized 'shiny products' over AI safetyFortune

Jan Leike resigned from OpenAI, citing concerns about the company's commitment to AI safety. His departure follows that of co-lead Ilya Sutskever, highlighting tensions within the organization about AI development.

β˜…β˜…β˜…β˜†β˜†Fullβ€”β€”safety1 articleView β†’
πŸ“„ PaperTopol: "High-performance medicine"Nature

Artificial intelligence, particularly deep learning, is revolutionizing healthcare by enhancing medical image interpretation, improving system workflows, and enabling personalized patient care through advanced data analysis.

β˜…β˜…β˜…β˜…β˜…FullTopol, Eric J.2019capabilitiesβ€”View β†’
πŸ”— WebTorβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— WebTow Center for Digital Journalismβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebToward Resisting AI-Enabled Authoritarianismβ€”Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ“„ PaperTowards Democratic AI GovernancearXivβ˜…β˜…β˜…β˜†β˜†SummaryShixiong Wang, Wei Dai, Geoffrey Ye Li2024-01-22governancedeep-learning+11 articleView β†’
πŸ“„ PaperTowards Guaranteed Safe AIarXivβ˜…β˜…β˜…β˜†β˜†SummaryDavid "davidad" Dalrymple, Joar Skalse, Yoshua Bengio, Stuart Russell, Max Tegmark, Sanjit Seshia, Steve Omohundro, Christian Szegedy, Ben Goldhaber, Nora Ammann, Alessandro Abate, Joe Halpern, Clark Barrett, Ding Zhao, Tan Zhi-Xuan, Jeannette Wing, Joshua Tenenbaum2024-05-10safetyinner-alignment+21 articleView β†’
πŸ”— WebTowards MonosemanticityTransformer Circuitsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebTracing the Roots of China's AI RegulationsCarnegie Endowmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governanceregulation+21 articleView β†’
✏️ Blog Posttracking by the Digital Forensic Research LabMediumβ˜…β˜…β˜†β˜†β˜†Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ“„ PaperTraining a Helpful and Harmless Assistant with RLHF (2022)arXivβ˜…β˜…β˜…β˜†β˜†SummaryYuntao Bai, Andy Jones, Kamal Ndousse, Amanda Askell, Anna Chen, Nova DasSarma, Dawn Drain, Stanislav Fort, Deep Ganguli, Tom Henighan, Nicholas Joseph, Saurav Kadavath, Jackson Kernion, Tom Conerly, Sheer El-Showk, Nelson Elhage, Zac Hatfield-Dodds, Danny Hernandez, Tristan Hume, Scott Johnston, Shauna Kravec, Liane Lovitt, Neel Nanda, Catherine Olsson, Dario Amodei, Tom Brown, Jack Clark, Sam McCandlish, Chris Olah, Ben Mann, Jared Kaplan2022-04-12alignmentgovernance+31 articleView β†’
πŸ“„ PaperTraining Language Models to Follow Instructions with Human FeedbackarXivβ˜…β˜…β˜…β˜†β˜†SummaryLong Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Christiano, Jan Leike, Ryan Lowe2022-03-04alignmentcapabilities+36 articlesView β†’
πŸ“„ PaperTransformer-XLarXivβ˜…β˜…β˜…β˜†β˜†SummaryZihang Dai, Zhilin Yang, Yiming Yang, Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov2019-01-09capabilitiesevaluation+31 articleView β†’
πŸ“„ PaperTree of ThoughtsarXivβ˜…β˜…β˜…β˜†β˜†SummaryShunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Thomas L. Griffiths, Yuan Cao, Karthik Narasimhan2023-05-17evaluationllm+31 articleView β†’
πŸ”— WebTrendForceβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebTrendForceβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebTrendForceβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebTrendForce ASML EUV analysis

ASML has established a near-monopoly in advanced semiconductor lithography equipment by developing EUV technology through extensive international partnerships and iterative innovation. Its competitive advantage stems from a sophisticated global supply chain and massive high-volume manufacturing data feedback loop.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ”— WebTrendForce Samsung 2nm

Samsung is emerging as a key 2nm chip manufacturer for Big Tech companies, leveraging TSMC's production limitations and geopolitical tensions.

β€”Fullβ€”β€”compute1 articleView β†’
πŸ”— WebTruepic

Truepic offers a digital verification platform that validates images, videos, and synthetic content using advanced metadata and detection technologies. The solution helps organizations prevent fraud and make more confident decisions across multiple industries.

β€”Fullβ€”β€”deepfakesdigital-evidence+22 articlesView β†’
πŸ”— WebTrump executive orderβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebTrust Research (Edelman)Edelmanβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”epistemiccascade+32 articlesView β†’
πŸ”— WebTruthfulQAGitHubβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”rlhfreward-hacking+12 articlesView β†’
πŸ“„ PaperTurner et al.arXivβ˜…β˜…β˜…β˜†β˜†SummaryAlexander Matt Turner2022-06-23alignmentpower-seeking+21 articleView β†’
πŸ”— WebTurner et al. (2021)β€”Noneβ€”β€”formal-analysispower-seeking+11 articleView β†’
πŸ“„ PaperTurner et al. formal resultsarXivβ˜…β˜…β˜…β˜†β˜†SummaryAlexander Matt Turner, Logan Smith, Rohin Shah, Andrew Critch, Prasad Tadepalli2019-12-03frameworkinstrumental-goals+36 articlesView β†’
πŸ”— WebTurner has expressed reservationsβ€”Noneβ€”β€”instrumental-convergenceself-preservation+11 articleView β†’
πŸ“„ PaperTurner's 2022 paperarXivβ˜…β˜…β˜…β˜†β˜†SummaryAlexander Matt Turner, Prasad Tadepalli2022-06-27governancesafety+32 articlesView β†’
πŸ”— WebTwitterβ€”Noneβ€”β€”iterated-amplificationscalable-oversight+11 articleView β†’
πŸ”— WebTwitter/Xβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebTwitter/X Transparency Reportsβ€”Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
πŸ“„ PaperTwo types of AI existential risk (2025)Springerβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebU.S. AI Safety InstituteBrookings Institutionβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety1 articleView β†’
πŸ›οΈ GovernmentU.S. Authorities Shut Down Major China-Linked AI Tech Smuggling Networkβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebU.S. Department of Defense Directive 3000.09β€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebU.S. Navy Close-In Weapons Systemβ€”Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— WebUC Berkeleyβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebUC Berkeleyβ€”Noneβ€”β€”adversarial-robustnessml-safety+11 articleView β†’
πŸ“„ PaperUC BerkeleyarXivβ˜…β˜…β˜…β˜†β˜†SummaryDavid Katona2023-06-16β€”1 articleView β†’
πŸ”— WebUC Berkeley Deepfake Researchβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebUC San Diego studyβ€”Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebUCL News: DeepMind co-founder and UCL alumnusβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentUK AI CouncilUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ›οΈ GovernmentUK AI Safety InstituteUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetyscaling+21 articleView β†’
πŸ›οΈ GovernmentUK AI Safety Institute renamed to AI Security InstituteUK AI Safety Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetycybersecurity+31 articleView β†’
πŸ“– ReferenceUK AI Safety Institute WikipediaWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”safety3 articlesView β†’
πŸ”— WebUK AI Safety Institute's Inspect frameworkβ€”Noneβ€”β€”safetyinterpretability+24 articlesView β†’
πŸ›οΈ GovernmentUK AI Security Institute's evaluationsUK AI Safety Instituteβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”evaluationcybersecurity2 articlesView β†’
πŸ›οΈ GovernmentUK AI White PaperUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”ai-forecastingcompute-trends+11 articleView β†’
πŸ›οΈ GovernmentUK AISIUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”capabilitythreshold+313 articlesView β†’
πŸ›οΈ GovernmentUK AISI/Gray Swan challengeUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentUK CMAUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ›οΈ GovernmentUK CMA investigationsUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ›οΈ GovernmentUK GovernmentUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”resource-allocationresearch-priorities+11 articleView β†’
πŸ›οΈ GovernmentUK Online SafetyUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safetysynthetic-media+21 articleView β†’
πŸ”— WebUK Parliamentβ€”Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ”— WebUkrainian autonomous drone swarmsβ€”Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— Webultra-low latency connections operating in the 300-800 nanosecond rangeβ€”Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ”— WebUNUnited Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ”— WebUN AI Advisory BodyUnited Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebUN Convention on Certain Conventional WeaponsUnited Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— WebUN Institute for Disarmament Researchβ€”Noneβ€”β€”escalationconflict+11 articleView β†’
πŸ”— WebUN Office for Disarmament AffairsUnited Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”timelineautomation+11 articleView β†’
πŸ”— WebUN Office of the High Representative for Disarmament Affairsβ€”Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ”— WebUN Secretary-General and ICRC issued a joint appealβ€”Noneβ€”β€”lawsmilitary-ai+11 articleView β†’
πŸ”— WebUN Security CouncilUnited Nationsβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”cybersecuritymonitoring+21 articleView β†’
πŸ”— WebUN Security Council Panel of Experts reportβ€”Noneβ€”β€”cybersecuritylaws+21 articleView β†’
πŸ”— WebUN Special Rapporteur on Privacyβ€”Noneβ€”β€”authoritarianismhuman-rights+11 articleView β†’
πŸ”— Webunanimous bipartisan supportβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperUnderspecification in Machine LearningarXivβ˜…β˜…β˜…β˜†β˜†SummaryAlexander D'Amour, Katherine Heller, Dan Moldovan, Ben Adlam, Babak Alipanahi, Alex Beutel, Christina Chen, Jonathan Deaton, Jacob Eisenstein, Matthew D. Hoffman, Farhad Hormozdiari, Neil Houlsby, Shaobo Hou, Ghassen Jerfel, Alan Karthikesalingam, Mario Lucic, Yian Ma, Cory McLean, Diana Mincu, Akinori Mitani, Andrea Montanari, Zachary Nado, Vivek Natarajan, Christopher Nielson, Thomas F. Osborne, Rajiv Raman, Kim Ramasamy, Rory Sayres, Jessica Schrouff, Martin Seneviratne, Shannon Sequeira, Harini Suresh, Victor Veitch, Max Vladymyrov, Xuezhi Wang, Kellie Webster, Steve Yadlowsky, Taedong Yun, Xiaohua Zhai, D. Sculley2020-11-06capabilitiestraining+11 articleView β†’
πŸ”— WebUnderstanding Mechanistic Interpretability in AI Models

Mechanistic interpretability is a technique for decoding how neural networks compute by analyzing their internal features, circuits, and computations. It seeks to translate complex model behaviors into human-understandable algorithms.

β€”Fullβ€”β€”interpretabilitycomputeβ€”View β†’
πŸ”— WebUnderstanding US Allies' Legal Authority on Export ControlsCSISβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”prioritizationtiming+12 articlesView β†’
πŸ”— WebUNESCO AI Ethicsβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— Webunfair trade practices under the Colorado Consumer Protection Actβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“‹ ReportUNICRIβ€”Noneβ€”β€”biosecuritydual-use-research+11 articleView β†’
πŸ”— WebUnited Nations Office for Disarmament Affairs (UNODA)β€”Noneβ€”β€”algorithmic-tradingfinancial-stability+11 articleView β†’
πŸ”— WebUniversity College Londonβ€”Noneβ€”β€”human-agencyautomation+11 articleView β†’
πŸ“„ PaperUniversity of MarylandarXivβ˜…β˜…β˜…β˜†β˜†SummarySeyed Mahed Mousavi, Simone Caldarella, Giuseppe Riccardi2023-05-25capabilitiestraining+31 articleView β†’
πŸ”— WebUniversity of Pennsylvania's Annenberg Schoolβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebUniversity of Torontoβ€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebUniversity of Washingtonβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— Webunprecedented growthβ€”Noneβ€”β€”chatgptgpt-4+11 articleView β†’
πŸ“„ PaperUnsolved Problems in ML SafetyarXivβ˜…β˜…β˜…β˜†β˜†SummaryDan Hendrycks, Nicholas Carlini, John Schulman, Jacob Steinhardt2021-09-28alignmentcapabilities+32 articlesView β†’
πŸ›οΈ GovernmentUpdated controls (October 2023)Bureau of Industry and Securityβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”internationalcompute-governance+11 articleView β†’
πŸ›οΈ GovernmentUS AI Safety and Security Boardβ€”Noneβ€”β€”safetycybersecurity+31 articleView β†’
πŸ›οΈ GovernmentUS AI Safety InstituteNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”safetygame-theory+35 articlesView β†’
πŸ›οΈ GovernmentUS AI Safety InstituteNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”safetymonitoring+33 articlesView β†’
πŸ›οΈ GovernmentUS AI Safety InstituteNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”safetyscaling+21 articleView β†’
πŸ›οΈ GovernmentUS AI Safety Institute vision documentNISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”safetyinner-alignment+21 articleView β†’
πŸ›οΈ GovernmentUS AISINISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentUS AISINISTβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”capability-generalizationalignment-stability+13 articlesView β†’
πŸ›οΈ GovernmentUS Department of Commerce - U.S. and UK Announce Partnership on Science of AI Safety

The US and UK have signed a Memorandum of Understanding to jointly develop AI safety tests and evaluations, focusing on information sharing and cooperative research between their respective AI Safety Institutes.

β€”Fullβ€”β€”safetyevaluation1 articleView β†’
πŸ›οΈ GovernmentUS DOJ Antitrustβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ›οΈ GovernmentUS Executive Order 14110β€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebUS Institute of Peaceβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentUS NAIRRβ€”Noneβ€”β€”ai-forecastingcompute-trends+11 articleView β†’
πŸ›οΈ GovernmentUS State Department - Strategic AI Partnership with Saudi Arabia

A bilateral agreement between the US and Saudi Arabia to collaborate on AI technologies, infrastructure development, and strategic investments across multiple sectors.

β€”Fullβ€”β€”β€”1 articleView β†’
πŸ›οΈ GovernmentUS-China Cooperation on AI SafetyCentre for the Governance of AIβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”safety1 articleView β†’
πŸ”— WebUSC Shoah Foundation

A nonprofit organization dedicated to recording, preserving, and sharing Holocaust survivor testimonies through innovative educational programs and digital platforms.

β€”Fullβ€”β€”historical-evidencearchives+11 articleView β†’
πŸ”— Webusing Constitutional AI to reduce sycophancyβ€”Noneβ€”β€”alignmenttruthfulness+11 articleView β†’
πŸ”— WebUyghur Human Rights Project (2024)β€”Noneβ€”β€”authoritarianismstability+11 articleView β†’
πŸ”— WebV-Dem Democracy Reportβ€”Noneβ€”β€”x-riskgovernance+11 articleView β†’
πŸ“„ PaperValue learningarXivβ˜…β˜…β˜…β˜†β˜†SummaryHiroshi Otomo, Bruce M. Boghosian, FranΓ§ois Dubois2017-11-09capabilities1 articleView β†’
πŸ“„ Papervan der Weij et al. (2024)arXivβ˜…β˜…β˜…β˜†β˜†SummaryTeun van der Weij, Felix HofstΓ€tter, Ollie Jaffe, Samuel F. Brown, Francis Rhys Ward2024-06-11governancecapabilities+31 articleView β†’
πŸ”— WebVaradi et al., NAR 2024β€”Noneβ€”β€”alphafolddrug-discovery+11 articleView β†’
πŸ›οΈ GovernmentVarious gov sourcesUK Governmentβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”x-riskeffective-altruism+11 articleView β†’
πŸ”— WebVasquez & Chen (2025). "Autonomous Cyber Operations: Capabilities and Limitations"β€”Noneβ€”β€”capabilitiescybersecurity+21 articleView β†’
πŸ”— WebVector Instituteβ€”Noneβ€”β€”deep-learningai-safety+11 articleView β†’
πŸ”— WebVeritone Q1 2025 Analysis

Analysis of U.S. labor market in Q1 2025 reveals significant growth in AI-related jobs, with 35,445 positions and a median salary of $156,998.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ›οΈ Governmentveto statementβ€”Noneβ€”β€”regulationstate-policy+33 articlesView β†’
πŸ”— WebVice: Replika Usersβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebVictoria Krakovnaβ€”Noneβ€”β€”capability-generalizationalignment-stability+11 articleView β†’
πŸ“„ PaperVillalobos et al.arXivβ˜…β˜…β˜…β˜†β˜†SummaryPablo Villalobos, Anson Ho, Jaime Sevilla, Tamay Besiroglu, Lennart Heim, Marius Hobbhahn2022-10-26capabilitiestraining+31 articleView β†’
πŸ”— WebVinod Khoslaβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebVirtasant

The article explores the economic value and implementation challenges of AI, highlighting potential cost savings and ROI considerations for enterprises adopting AI technologies.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ“„ PaperVisibility into AI ChipsarXivβ˜…β˜…β˜…β˜†β˜†SummaryYonadav Shavit2023-03-20capabilitiestraining+11 articleView β†’
πŸ”— WebVision Language Models 2025 - Hugging Face

This comprehensive review explores the latest developments in Vision Language Models, highlighting innovations in model architectures, reasoning capabilities, and specialized applications like robotics and multimodal agents.

β€”Fullβ€”β€”capabilitiesllmβ€”View β†’
✏️ Blog PostVision, Voice, and Beyond: Multimodal AI in 2025

Multimodal AI models can interpret and generate content across different media types, enabling complex interactions like image-based recipe suggestions and real-time translation. These models represent a significant advancement in AI's ability to understand and communicate.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebVitalik Buterin PhD FellowshipFuture of Life Instituteβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebVoice cloning now requires just 3 seconds of audioβ€”Noneβ€”β€”social-engineeringvoice-cloning+11 articleView β†’
πŸ”— WebVoice cloning with 3 seconds of audioβ€”Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
πŸ”— Webvoided non-disparagement termsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— Webvoluntary commitments only lead to socially beneficial outcomes when combined with enforcement mechanismsScienceDirectβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebVox: "AI Chatbots Will Tell You What You Want to Hear"β€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebvTaiwanβ€”Noneβ€”β€”democratic-innovationcollective-intelligence+11 articleView β†’
πŸ”— WebvTaiwan case study

vTaiwan is a digital democracy platform that uses Pol.is and AI to gather public input on complex policy issues, enabling collaborative and consensual decision-making in Taiwan across various domains.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ“„ PaperWAICONatureβ˜…β˜…β˜…β˜…β˜…Noneβ€”β€”regulationchina+12 articlesView β†’
πŸ”— WebWald AI: ChatGPT Data Leaks and Security Incidents

A comprehensive review of ChatGPT security incidents reveals numerous data breaches, credential thefts, and privacy concerns from 2023 to 2025. The incidents highlight critical challenges in AI data protection and user privacy.

β€”Fullβ€”β€”cybersecurityβ€”View β†’
πŸ“„ PaperWang et al. (2021)arXivβ˜…β˜…β˜…β˜†β˜†SummaryDouwe Kiela, Max Bartolo, Yixin Nie, Divyansh Kaushik, Atticus Geiger, Zhengxuan Wu, Bertie Vidgen, Grusha Prasad, Amanpreet Singh, Pratik Ringshia, Zhiyi Ma, Tristan Thrush, Sebastian Riedel, Zeerak Waseem, Pontus Stenetorp, Robin Jia, Mohit Bansal, Christopher Potts, Adina Williams2021-04-07capabilitiesevaluation+31 articleView β†’
πŸ”— WebWang et al. (2024)β€”Noneβ€”β€”instrumental-convergenceself-preservation+11 articleView β†’
πŸ”— WebWarns of "crucial considerations"β€”Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ“„ PaperWatermarking language modelsarXiv

Researchers propose a watermarking framework that can embed signals into language model outputs to detect machine-generated text. The watermark is computationally detectable but invisible to humans.

β˜…β˜…β˜…β˜†β˜†FullKirchenbauer, John, Geiping, Jonas, Wen, Yuxin, Katz, Jonathan, Miers, Ian, Goldstein, Tom2024llmdeepfakes+21 articleView β†’
πŸ”— WebWCCFtechβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebWCCFtechβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebWCCFtech Huawei capacity

A CSIS report suggests Huawei has found ways to circumvent US chip sanctions by acquiring manufacturing equipment and stockpiling previous-generation chip dies. The company may produce around 750,000 Ascend 910C AI chips through creative manufacturing approaches.

β€”Fullβ€”β€”compute1 articleView β†’
πŸ›οΈ GovernmentWeak voluntaryWhite Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”risk-factorgame-theory+32 articlesView β†’
πŸ”— WebWeak-to-strong generalizationOpenAI

A research approach investigating weak-to-strong generalization, demonstrating how a less capable model can guide a more powerful AI model's behavior and alignment.

β˜…β˜…β˜…β˜…β˜†Fullβ€”β€”alignmenttraining+35 articlesView β†’
πŸ”— WebWeapons of Mass Destruction Proxy Benchmark (WMDP)β€”Noneβ€”β€”capabilitiesevaluation1 articleView β†’
πŸ”— WebWebArenaβ€”Noneβ€”β€”agenticplanning+32 articlesView β†’
πŸ”— WebWebProNews, 2025β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebWEF Future of Jobs 2025World Economic Forumβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economiclabor-markets+21 articleView β†’
πŸ”— WebWEF Global Risks Report 2024World Economic Forumβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebWEF projectsWorld Economic Forumβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”labor-marketsautomation+11 articleView β†’
πŸ“„ PaperWei et al. (2023)arXivβ˜…β˜…β˜…β˜†β˜†SummaryJames Waldron, Leon Deryck Loveridge2023-11-15rlhfreward-hacking+11 articleView β†’
πŸ“„ PaperWei et al. (2023): "Simple Synthetic Data"arXivβ˜…β˜…β˜…β˜†β˜†SummaryJerry Wei, Da Huang, Yifeng Lu, Denny Zhou, Quoc V. Le2023-08-07capabilitiesevaluation+1β€”View β†’
πŸ”— WebWen et al. 2024β€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebWest Point's Modern War Instituteβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebWestern nations worry that UN involvement could open the door to Chinese and autocratic influenceβ€”Noneβ€”β€”β€”1 articleView β†’
✏️ Blog PostWhat Failure Looks LikeAlignment Forumβ˜…β˜…β˜…β˜†β˜†Metadatapaulfchristiano2019-03-17iterated-amplificationscalable-oversight+13 articlesView β†’
πŸ”— WebWhat-If Tool (Google)β€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebWhat's changed since the "pause AI" letter six months ago?MIT Technology Reviewβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”β€”2 articlesView β†’
πŸ”— WebWhich? investigationβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebWhite & Case Global AI Regulatory Trackerβ€”Noneβ€”β€”governance1 articleView β†’
πŸ”— WebWhite & Case: AI Watch Global Regulatory Tracker

White & Case's global AI regulatory tracker highlights the complex and inconsistent approaches different countries are taking to AI regulation. The analysis reveals significant variations in legal frameworks, definitions, and enforcement strategies.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ”— WebWhite & Case: EU AI Act Becomes Law

The EU AI Act is a pioneering regulation establishing comprehensive rules for AI development, deployment, and use. It introduces a risk-based approach with significant penalties for non-compliance.

β€”Fullβ€”β€”governanceβ€”View β†’
πŸ›οΈ GovernmentWhite House AI Bill of RightsWhite Houseβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ›οΈ GovernmentWhite House AI commitmentsβ€”Noneβ€”β€”self-regulationindustry-commitments+12 articlesView β†’
πŸ”— WebWHO's Disease Outbreak Newsβ€”Noneβ€”β€”probabilitydecomposition+11 articleView β†’
✏️ Blog PostWhy AI X-Risk Skepticism?LessWrongβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”x-risk1 articleView β†’
πŸ”— WebWickens et al. (2015)β€”Noneβ€”β€”human-ai-interactionai-control+11 articleView β†’
✏️ Blog PostWidening AI Safety's Talent PipelineEA Forumβ˜…β˜…β˜…β˜†β˜†MetadataRubenCastaing, Nelson_GC, danwil2025-09-25safetyfield-building+21 articleView β†’
πŸ“– ReferenceWikidataWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ”— WebWikidataβ€”Noneβ€”β€”market-concentrationgovernance+11 articleView β†’
πŸ”— WebWikimedia Foundation

The Wikimedia Foundation hosts Wikipedia, a nonprofit-driven encyclopedic platform with over 65 million articles across 300+ languages. It relies on nearly 265,000 monthly volunteers to create and maintain reliable, open-access information.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebWikimedia research

Wikimedia Research aims to advance understanding of Wikimedia projects by conducting research, developing technologies, and supporting communities through scientific approaches.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ“– ReferenceWikipediaWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”knowledge-managementpublic-goods+11 articleView β†’
πŸ“– ReferenceWikipedia's accountWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”chatgptgpt-4+11 articleView β†’
πŸ”— WebWILDS benchmarkβ€”Noneβ€”β€”capabilitiesevaluation+31 articleView β†’
πŸ“„ PaperWill Compute Bottlenecks Prevent an Intelligence Explosion?arXivβ˜…β˜…β˜…β˜†β˜†SummaryParker Whitfill, Cheryl Wu2025-07-31compute1 articleView β†’
πŸ”— Webwilsoncenter.org/program/science-and-technology-innovation-programβ€”Noneβ€”β€”historical-evidencearchives+11 articleView β†’
πŸ”— WebWired AI Coverageβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebWired: AI Companionsβ€”Noneβ€”β€”mental-healthai-ethics+11 articleView β†’
πŸ”— WebWired: Reality Splitβ€”Noneβ€”β€”β€”β€”View β†’
πŸ”— WebWired: The End of Trustβ€”Noneβ€”β€”deepfakesdigital-evidence+11 articleView β†’
✏️ Blog PostWithout specific countermeasures, the easiest path to transformative AI likely leads to AI takeoverAlignment Forumβ˜…β˜…β˜…β˜†β˜†MetadataAjeya Cotra2022-07-18β€”1 articleView β†’
πŸ”— WebWitness

A global organization that trains and supports human rights defenders in using video and technology to capture and preserve evidence of violations. Focuses on countering potential AI-generated misinformation.

β€”Fullβ€”β€”deepfakesdigital-evidence+32 articlesView β†’
πŸ”— WebWITNESS Media Lab

A multimedia project focusing on using citizen-generated video to expose human rights abuses and develop technological strategies for video verification and justice.

β€”Fullβ€”β€”historical-evidencearchives+11 articleView β†’
πŸ”— WebWitness: "Ticks or It Didn't Happen"

A multi-disciplinary initiative focused on preparing for potential malicious uses of AI-generated synthetic media, emphasizing global human rights and inclusive solutions.

β€”Fullβ€”β€”β€”β€”View β†’
πŸ”— WebWolfers & Zitzewitz (2004)

Wolfers & Zitzewitz analyze prediction markets as a method for efficiently aggregating information and generating forecasts across various domains, demonstrating their accuracy and potential utility.

β€”Fullβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebWolfers & Zitzewitz (2006)β€”Noneβ€”β€”forecastinginformation-aggregation+11 articleView β†’
πŸ”— WebWork at MIRIMIRIβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”x-riskirreversibility+11 articleView β†’
πŸ”— WebWorld Bank WGI 2024

The World Bank's Worldwide Governance Indicators (WGI) measure six key governance dimensions using perception data from multiple sources. The 2025 edition introduces methodological updates to improve cross-country governance comparisons.

β€”Fullβ€”β€”governance1 articleView β†’
πŸ”— WebWorld Economic ForumWorld Economic Forumβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic1 articleView β†’
πŸ”— WebWorld Economic Forum

A comprehensive analysis of AI's impact on jobs, skills, and wages across six continents, showing positive transformative effects rather than job displacement.

β€”Fullβ€”β€”economic1 articleView β†’
πŸ”— WebWorld Economic ForumWorld Economic Forumβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic1 articleView β†’
πŸ”— WebWorld Economic ForumWorld Economic Forumβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economictimeline+21 articleView β†’
πŸ”— WebWorld Economic ForumWorld Economic Forumβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economicdisinformation+21 articleView β†’
πŸ”— WebWorld Economic ForumWorld Economic Forumβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic1 articleView β†’
πŸ”— WebWorld Economic Forum notesWorld Economic Forumβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”economic1 articleView β†’
πŸ”— WebWorld Economic Forum: Governance in the Age of Generative AIWorld Economic Forumβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”governanceeconomic1 articleView β†’
πŸ”— WebWorld Economic Forum's 2024 Global Risks Reportβ€”Noneβ€”β€”economic1 articleView β†’
πŸ”— WebWorld Economic Forum's 2024 white paper on AI Value AlignmentWorld Economic Forumβ˜…β˜…β˜…β˜…β˜†Noneβ€”β€”alignmenteconomic1 articleView β†’
πŸ”— WebWorld Privacy Forum's technical analysisβ€”Noneβ€”β€”deepfakesdigital-evidence+33 articlesView β†’
πŸ”— WebWSJ Facebook Filesβ€”Noneβ€”β€”human-agencyautonomy+11 articleView β†’
πŸ”— WebWSJ: Facebook Filesβ€”Metadataβ€”2021ai-ethicspersuasion+11 articleView β†’
πŸ”— WebX (formerly Twitter) under Elon Muskβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebX/Twitterβ€”Noneβ€”β€”gpt-4chatgpt+11 articleView β†’
πŸ”— WebxAIβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebXPT Results

The Existential Risk Persuasion Tournament gathered 169 participants to forecast potential human extinction risks by 2100, examining perspectives on AI, nuclear war, pandemics, and other global threats.

β€”Fullβ€”β€”x-riskβ€”View β†’
πŸ”— WebYale Climate Opinionβ€”Noneβ€”β€”filter-bubblespolarization+11 articleView β†’
πŸ”— WebYale Program on Climate Changeβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebYale's Fiona Scott Mortonβ€”Noneβ€”β€”governancepower-dynamics+11 articleView β†’
πŸ”— WebYale's Social Cognition and Decision Sciences Labβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebYann LeCunβ€”Noneβ€”β€”formal-analysispower-seeking+11 articleView β†’
πŸ“– ReferenceYann LeCunWikipediaβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebYann LeCun's posts

I apologize, but the provided content appears to be an error page from X (formerly Twitter) and does not contain any substantive text from Yann LeCun's posts. Without the actual content of his posts, I cannot generate a meaningful summary. To properly analyze Yann LeCun's posts, I would need: 1. The specific text of his posts 2. Context about the topic he was discussing 3. The source and date of the posts If you can provide the actual content of the posts, I'll be happy to create a comprehensive summary following the requested JSON format. Would you like to: - Recheck the source document - Provide the posts in text form - Choose a different source to analyze

β€”Summaryβ€”β€”eliciting-latent-knowledgeelk+11 articleView β†’
πŸ”— WebYoshua Bengioβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ“„ PaperYoshua Bengio and othersarXivβ˜…β˜…β˜…β˜†β˜†SummaryYoshua Bengio, Geoffrey Hinton, Andrew Yao, Dawn Song, Pieter Abbeel, Trevor Darrell, Yuval Noah Harari, Ya-Qin Zhang, Lan Xue, Shai Shalev-Shwartz, Gillian Hadfield, Jeff Clune, Tegan Maharaj, Frank Hutter, AtΔ±lΔ±m GΓΌneş Baydin, Sheila McIlraith, Qiqi Gao, Ashwin Acharya, David Krueger, Anca Dragan, Philip Torr, Stuart Russell, Daniel Kahneman, Jan Brauner, SΓΆren Mindermann2023-10-26governancecapabilities+31 articleView β†’
πŸ”— WebYouGov

A recent YouGov survey shows increasing American concerns about AI, with 43% worried about potential human extinction and 47% believing AI's societal effects will be negative.

β€”Fullβ€”β€”x-risk1 articleView β†’
🎀 TalkYouTubeβ€”Noneβ€”β€”adversarial-robustnessml-safety+11 articleView β†’
🎀 TalkYouTube Channelsβ€”Noneβ€”β€”β€”1 articleView β†’
πŸ”— WebYouTube's approach to synthetic mediaβ€”Noneβ€”β€”disinformationinfluence-operations+11 articleView β†’
πŸ”— WebZebracat/DemandSageβ€”Noneβ€”β€”labor-marketsautomation+11 articleView β†’
πŸ“„ PaperZhang et al. (2021)arXivβ˜…β˜…β˜…β˜†β˜†SummaryAbeba Birhane, Pratyusha Kalluri, Dallas Card, William Agnew, Ravit Dotan, Michelle Bao2021-06-29interpretabilitycapabilities+31 articleView β†’
πŸ“„ PaperZiegler et al. (2022)arXivβ˜…β˜…β˜…β˜†β˜†SummaryMaxime Darrin, Pablo Piantanida, Pierre Colombo2022-12-18alignmentinterpretability+31 articleView β†’
πŸ“„ PaperZKML Survey (Kang et al.)arXivβ˜…β˜…β˜…β˜†β˜†SummarySean J. Wang, Honghao Zhu, Aaron M. Johnson2023-10-12capabilitiessafety+31 articleView β†’
πŸ“„ PaperZou et al. (2024): Forecasting Future World Events with Neural NetworksarXivβ˜…β˜…β˜…β˜†β˜†SummaryAnkit Khandelwal, Handy Kurniawan, Shraddha Aangiras, Γ–zlem Salehi, Adam Glos2023-12-18forecastingprediction-markets+11 articleView β†’
πŸ”— WebZuckerberg signaledTechCrunchβ˜…β˜…β˜…β˜†β˜†Noneβ€”β€”β€”1 articleView β†’