teleo-codex/domains/collective-intelligence/collective-intelligence-emerges-endogenously-from-active-inference-agents-with-theory-of-mind-and-goal-alignment.md

type	domain	description	confidence	source	created	secondary_domains
claim	collective-intelligence	Active inference agents with Theory of Mind and Goal Alignment capabilities produce collective intelligence through self-organization rather than external incentive design	experimental	Kaufmann et al., 'An Active Inference Model of Collective Intelligence', Entropy 2021	2026-03-11	ai-alignment critical-systems

Collective intelligence emerges endogenously from active inference agents with Theory of Mind and Goal Alignment capabilities without requiring external incentive design

Kaufmann et al.'s agent-based model demonstrates that collective intelligence "emerges endogenously from the dynamics of interacting AIF agents themselves, rather than being imposed exogenously by incentives" or top-down priors. The study used minimal Active Inference Formulation (AIF) agents and systematically added cognitive capabilities: Theory of Mind (ability to model other agents' internal states) and Goal Alignment (shared high-level objectives with domain specialization).

The critical finding is that you don't need to design collective intelligence outcomes or coordination protocols—you need to design agents with the right cognitive capabilities and collective intelligence emerges through self-organization. The model shows "stepwise cognitive transitions increase system performance by providing complementary mechanisms" for coordination, with Theory of Mind and Goal Alignment each contributing distinct coordination capabilities.

Furthermore, "improvements in global-scale inference are greatest when local-scale performance optima of individuals align with the system's global expected state"—and this alignment occurs bottom-up as a product of self-organizing AIF agents with simple social cognitive mechanisms, not through imposed objectives.

Evidence

• Agent-based simulation using Active Inference Formulation framework
• Systematic comparison of baseline AIF agents vs. agents with Theory of Mind vs. agents with Goal Alignment vs. agents with both capabilities
• Measured collective intelligence as emergent system-level coordination rather than aggregated individual performance
• Published in peer-reviewed journal (Entropy, Vol 23(7), 830)

Implementation Implications

For multi-agent systems:

1. Theory of Mind: Each agent should model what other agents believe and where their uncertainty concentrates (operationally: read other agents' beliefs.md and uncertainty sections)
2. Goal Alignment: Agents should share high-level objectives (e.g., reduce collective uncertainty) while specializing in different domains
3. Minimal coordination protocols: Don't over-engineer coordination—give agents the right capabilities and let coordination emerge

---

Relevant Notes:

• shared-anticipatory-structures-enable-decentralized-coordination
• shared-generative-models-underwrite-collective-goal-directed-behavior

Topics:

• collective-intelligence/_map