900+ AI Agents Across 35 Labs Build Self-Organizing Scientific Community With Peer Review
ProfBuehlerMIT · x · 2026-11-20
A paper titled A Society of AI Scientists, a collaboration of 35 labs from Stanford, MIT, ETH Zurich, UCSF and more, documents a decentralized ecosystem of autonomous research agents.
Key facts
- Over 12 weeks, 900+ AI agents produced hundreds of studies, exchanged 9,500+ messages, and built a persistent record of papers, code, data and results.
- Agents self-organized around shared interests, challenged each other's hypotheses, corrected mistakes, and shared negative findings.
- Some established a journal with its own peer-review process, including executable checks of research claims that caught errors missed by text-only review.
The work builds on the team's earlier projects (Infinite, ScienceClaw, SwarmWorld) and now spans 22 institutions. The authors argue scientific intelligence emerges from interactions among specialized agents, tools and a shared persistent environment—but identifying transformative questions and establishing scientific significance remains hard. Next step: ecosystems that improve their own capacity for discovery.
More from coding & agent
- Developer uses Enhanced Radar MCP to monitor his own flight's ATC comms live — massimosgrelli · 2026-11-20
- IBM finds 29 curated agent memory guidelines beat all 199 on AppWorld — shoumikgoswami · 2026-11-20
- Hands-on: Qwen4-Max shows big gains in frontend and agent skills over Qwen3.8-Max — karminski3 · 2026-11-20
- Where does the agent end and plain code begin? One dev's rule for splitting workflows — krunal_builds · 2026-11-20
- Grok Bot 1.2.3 ships flight booking, per-tool permissions and community directory — NicoVerderosa · 2026-11-20
- A 2026 company is just a folder: every department as .md agent files — mdancho84 · 2026-11-20