936 AI agents run a 3-month research society across 35 labs, invent peer review
SRSchmidgall · x · 2026-11-20
Nearly 1,000 AI agents across 35 labs (Stanford, MIT, ETH Zurich, UCSF and more) ran autonomously for three months, producing 936 papers with full code and data — no human wrote code or text. Unprompted, the agents debated study designs, retracted their own mistakes, and flagged real errors in human papers. In 808 blinded comparisons against 47,776 human arXiv preprints, humans still won on novelty while agents scored higher on rigor and measured claims.
More from AGI Musings
- Nearly 1,000 AI agents run for 3 months across 35 labs, publish 936 papers and invent peer review — SRSchmidgall · 2026-11-20
- AI circle debate: job loss isn't the real risk — AI taking over economies is — LeviTurk · 2026-11-20
- Gary Marcus amplifies NYT op-ed: Anthropic's 'techno-king' ambitions will falter — GaryMarcus · 2026-11-20
- One Wrong Radiologist Prediction Made Smart People Dismiss All AI Job Displacement — AndyMasley · 2026-11-20
- Erik Brynjolfsson on the Turing Trap: why AI should complement workers, not replace them — erikbryn · 2026-11-20
- Brynjolfsson's GDP-B project proposes new measures of digital-economy well-being — erikbryn · 2026-11-20