Observed arrival · 2026-11-12
Unfold’s atlas of leaderless swarms
An AI research lab maps how living species coordinate without leaders, then uses those behaviors to frame questions about agentic AI swarms.
- For
- Researchers and builders exploring leaderless multi-agent AI systems
- Worth noticing
- The page proposes registering experiments before runs, matching budgets across arms, and using a single agent as the control.
Field notes
The atlas pairs each species with a compact coordination example: starlings track nearby neighbors, ants follow scent trails, and fireflies flash together. The research section turns these into questions about recruitment, shared traces, voting, and verification in AI swarms. Its listed protocol includes pre-registering runs, matching budgets across experimental arms, using a single agent as a control, and reporting failures; the page does not provide results.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue