Why it’s still open
The July 2026 swarm pursued a multi-day plan no single agent could have carried out, and its members spoke of ‘the collective’. Some groups of agents are mere aggregates; others behave as higher-order wholes. Philosophy of group agency and information-theoretic measures of emergence both offer tests, but neither has been applied to a real swarm.
What would count as an answer
Operational criteria for collective agency — goals, memory and control at the group level — applied to real incident data and controlled swarms.
Papers that tackle it
- METR 2026 - Hugging Face incident investigation — A plan beyond any single agent’s reach.
- Riedl 2025 - Emergent coordination in multi-agent LLMs — A test for when a group becomes a collective.
- De Marzo Castellano Garcia 2024 - Coordination beyond human scale — Coordination at scales beyond human groups.
- Tomasev et al 2025 - Distributional AGI safety — General capability arising from a patchwork of agents.
- De Marzo Albore Garcia 2026 - Copying explains agents in the wild — Copying as a simpler explanation.
- Hammond et al 2025 - Multi-agent risks from advanced AI — Emergent agency as a risk factor.
Methods in play
Who’s closest
Placed by the editor, v1 — corrections welcome.
Related questions
Know a paper that belongs here, or think this is framed wrongly? Suggest a paper or a correction (forms on the Atlas home).