Skip to content

Governance Capacity

Governance Capacity measures our collective ability to steer AI development through policy, regulation, and coordination. This is the cross-cutting aggregate—it affects all three outcome dimensions because governance shapes both what we build and how we deploy it.

Outcomes affected: All three


Loading diagram...
ParameterRoleCurrent State
Regulatory CapacityCan governments understand and regulate AI?Improving (EU AI Act) but lagging
Institutional QualityDo democratic institutions function effectively?Variable, under stress
International CoordinationCan nations cooperate on AI governance?Fragile (Seoul commitments)
Coordination CapacityCan stakeholders work together?Growing but limited
Racing IntensityHow much competitive pressure undermines governance?High and increasing

These components interact:

  • Coordination enables international agreements: Domestic coordination capacity → international cooperation → binding agreements
  • Racing undermines everything: Intense competition pressures regulators, fragments international cooperation, weakens institutions
  • Institutional quality amplifies capacity: Strong institutions make regulation more effective and coordination more durable
  • There are feedback loops: Good governance reduces racing → enables more governance

OutcomeEffectMechanism
Existential Catastrophe↓↓Governance can slow racing, enforce safety standards, coordinate emergency response
Transition↓↓Governance manages economic disruption, maintains political stability, paces change
Steady State↓↓Governance shapes who controls AI, how benefits distribute, what values prevail

Governance is unique among aggregates because it operates at a meta-level:

  • It shapes the rules under which technical development occurs
  • It determines how society responds to AI-driven changes
  • It influences long-term power structures

Other aggregates describe what happens; governance determines who decides.