Google's ScientistTwo AI writes expert-level papers end-to-end and self-reviews
2C_ornot2C · x · 2026-11-09
Google Cloud AI Research and University of Waterloo introduce ScientistTwo, a fully autonomous multi-agent framework that takes a problem, sets SOTA baselines, generates hypotheses, runs experiments, debugs code, and validates findings via a closed-loop simulated peer-review rebuttal engine. Benchmarked against papers accepted at ICLR/ICML/NeurIPS, it autonomously produces expert-level publishable papers and verified codebases that outperform human SOTA models and score higher than human-authored papers under AI reviewers. Project page: scientist-two.github.io.
More from AGI Musings
- Silicon Valley warns AI will kill everyone — normies respond by voting against data centers — DavidSKrueger · 2026-11-09
- Banning AI Is Like Banning Uranium Enrichment, Safety Researchers Quip — DavidSKrueger · 2026-11-09
- AI patronage model questioned: keeping access internal is just as disempowering — andersonbcdefg · 2026-11-09
- Cambridge AI safety prof delighted by Terminator one-liner used to pitch AI risk to Americans — DavidSKrueger · 2026-11-09
- Ending AI labs' power to pick winners via early model access, argued as key anti-RSI lever — andersonbcdefg · 2026-11-09
- AI safety debate: would a 50-year delay of frontier AI even help? — ohabryka · 2026-11-09