AGI Strategies
← overview

Cooperation substrate ↑ · ai artefact

Cooperative AI

The binding constraint is equilibrium dynamics of many AI systems, not individual alignment; commitment and verification tech make cooperative equilibria reachable.

Mechanism

Design AI systems, protocols, and institutions for stable cooperative equilibria between AI systems.

If it succeeds: what binds next

AI agents cooperate. The binding problem becomes exclusion from the cooperating group, which humans, which AIs, which goals are in.

A strategy that produces a worse next problem than the one it solved has not done durable work.

Falsification signal

AI systems defect in deployments where commitment technology exists and cooperation was available.

A strategy held without a falsification signal is not strategy; it is affiliation. Continued support after this signal lands is identity, not bet. See the identity diagnostic.

Self-undermining threshold

overshoot risk

Before commitment and verification technology is mature enough for stable cooperation.

Unstable multi-agent regimes inherit the failure modes of imperfect cooperation without the benefits of robust cooperation.

Every strategy has a stable region where it reinforces itself and an unstable region where pursuit defeats it. The threshold between them is usually narrower than advocates acknowledge.

People on the record

6

Profiled figures appear first, with their tier in small caps. Each face links to the person and their full quote record. Tag: cooperative-ai.

  • Andrew Critch

    Deep ML / safety technical · Recognised inside subfield

  • Vincent Conitzer

    Deep ML / safety technical · Recognised inside subfield

  • Igor Mordatch

    Google DeepMind; multi-agent and embodied AI

  • Jakob Foerster

    Jakob Foerster

    Oxford FLAIR lab; multi-agent RL

  • Kate Sills

    AI economic systems and multi-agent markets researcher

  • Lewis Hammond

    Cooperative AI Foundation co-director

Load-bearing commitments

Worldview positions this strategy quietly assumes. If the claim fails empirically or philosophically, the strategy loses its target or its premise.

Agency

AI has sufficient agency for reciprocal arrangement.

Fails if: If AI agency is instrumental, cooperation is a design choice humans make, not a two-sided arrangement.

Coordinates

Primary leverCooperation substrate (AI-inclusive)
Acts onai artefact
Coercionconsent
Actor in controlmulti ai
Time horizonduring transition
Legitimacy sourcetechnical

Conflicts, grouped by mechanism

0

No strict conflicts catalogued. This strategy pulls a lever that nothing else pulls in the opposite direction.

Complements, grouped by mechanism

5

Same-lever reinforce

same lever, same pull, different mechanism

Both strategies pull the same lever in the same direction by different means. They stack: doing both amplifies the pull, at the cost of double-counting in portfolio audits.

Coordination infrastructureAI welfare as safetyMutual dependency

Same phase, different layer

same stage, distinct levers

Both are active in the same phase of the transition but act on different layers (model vs institution vs culture). They cover different failure modes inside the same window.

MultipolarityPlural AI ethic

Axis position

What the strategy acts onAI artefact
Coercion levelConsent
Actor in controlMulti-AI equilibrium
Time horizonDuring transition
Legitimacy sourceTechnical

Source note: Cooperative AI strategy.md