Skip to content

Search stories, models, companies…Subscribe

Topic · Concept

AI for science

AI for science is the use of AI models and agents in research, from designing proteins and drug molecules to proving theorems and analyzing physics data. The claims move fast and are often made by the companies selling the models, so the useful questions are what was checked, by whom, and what still needs lab work or peer review.

6 stories · updated Oct 24

Follow via RSS Get AI for science alerts

Key facts

Checked Oct 23

2024 Nobel Prize in Chemistry: Announced October 9, 2024: shared by Demis Hassabis and John Jumper (Google DeepMind) for AlphaFold, and David Baker for computational protein design
Genesis Mission: Executive Order 14363, signed November 24, 2025 (90 FR 55035); implemented by the Secretary of Energy through the American Science and Security Platform; initial operating capability targeted within 270 days, subject to appropriations
Navier–Stokes: Clay Mathematics Institute: September 11, 2026: CMI says the problem "has apparently been settled" and that its evaluation is "deliberately unhurried"; no prize decision announced
Navier–Stokes: OpenAI's Lean certificates: OpenAI's public repository holds Lean 4 certificates for its Navier–Stokes and Euler results
KLS conjecture: Microsoft paper: "An O(1) Bound for the KLS Constant," Zhao Song and Xinzhi Zhang; arXiv 2610.01447, v1 October 1, 2026, v2 October 4, 2026
KLS conjecture: independent proof: "Presenting a proof of the Kannan-Lovasz-Simonovits conjecture," Pierre Bizeul, Boaz Klartag and Joseph Lehec; arXiv 2610.05474, October 4, 2026
KLS conjecture: AI's role: Microsoft Research, October 21, 2026: AI agents explored about one hundred research trajectories and helped with derivations, drafting and internal proof checks; the researchers chose the direction, assembled the argument and verified the proof
Claude scorpion toxin binders: Anthropic, October 20, 2026: all 211 designs lab-tested by Adaptyv Bio and Twist Bioscience; 17 met the binding criteria, against four of five target toxins; Anthropic says it does "not claim to have produced an antivenom"
CERN and Mistral AI: October 15, 2026: statement of intent for a long-term partnership, including a pretrained "Large Physics Model" for collider physics trained on simulated Large Hadron Collider collisions
Warp Sciences: Describes itself as a US chemistry contract research organization (CRO) that builds "automated labs that learn from every experiment"; Kevin Weil is CEO; South San Francisco, California
LLM-assisted peer review: "Beyond Imitation: A Framework and Benchmark for LLM-Assisted Peer Review," Rachel S.Y. Teo et al.; arXiv 2610.11087, October 8, 2026

Start here

Industry · NewsWarp Sciences raises $215 million: Kevin Weil's AI-and-robotics chemistry lab wants to bring drug-development work back from ChinaWarp Sciences, Kevin Weil's startup after OpenAI, came out of stealth on October 22, 2026, with a $215 million Series A led by Lightspeed. What it builds, who backs it, and the China outsourcing and BIOSECURE Act context.Oct 24·5 min read

Research · AnalysisClaude scorpion toxin binders: Anthropic says 17 of 211 lab-tested designs met its criteria, but none is an antivenomAnthropic reports that Claude designed candidate protein binders against scorpion toxins in a 64-hour campaign. Two contract labs tested 211 designs; 17 met Anthropic's binding criteria, each against a single toxin. None has been tested for neutralizing venom.Oct 24·5 min read

Research · NewsKLS conjecture proof: three papers claim the 31-year-old problem, and what AI did in eachThe KLS conjecture proof arrived in early October 2026 from more than one team. Microsoft's Xinzhi Zhang and Zhao Song credit a human-directed AI agent search; Bizeul, Klartag and Lehec say most of their proofs were found by ChatGPT. None is peer reviewed yet.Oct 24·6 min read

Latest on AI for science

Industry · NewsWarp Sciences raises $215 million: Kevin Weil's AI-and-robotics chemistry lab wants to bring drug-development work back from ChinaWarp Sciences, Kevin Weil's startup after OpenAI, came out of stealth on October 22, 2026, with a $215 million Series A led by Lightspeed. What it builds, who backs it, and the China outsourcing and BIOSECURE Act context.Oct 24·5 min read

Research · NewsKLS conjecture proof: three papers claim the 31-year-old problem, and what AI did in eachThe KLS conjecture proof arrived in early October 2026 from more than one team. Microsoft's Xinzhi Zhang and Zhao Song credit a human-directed AI agent search; Bizeul, Klartag and Lehec say most of their proofs were found by ChatGPT. None is peer reviewed yet.Oct 24·6 min read

Research · AnalysisClaude scorpion toxin binders: Anthropic says 17 of 211 lab-tested designs met its criteria, but none is an antivenomAnthropic reports that Claude designed candidate protein binders against scorpion toxins in a 64-hour campaign. Two contract labs tested 211 designs; 17 met Anthropic's binding criteria, each against a single toxin. None has been tested for neutralizing venom.Oct 24·5 min read

Open source · NewsMistral Timestral: four open-weight time-series models, a $20M revenue cap, and a CERN statement of intentMistral Timestral is a family of four time-series forecasting models, from 5M to 209M parameters, with weights on Hugging Face under a revenue-capped 'Modified MIT' license. The same week, CERN and Mistral signed a statement of intent, not a contract, to build a Large Physics Model.Oct 20·5 min read

Research · AnalysisSakana AI's Multi-Layered Review: a Claude-based peer-review system that finds planted errors, and where it falls shortSakana AI's Multi-Layered Review (MLR), described in a TMLR paper, uses three Claude agents to review ML papers. On a new benchmark of 1,164 planted contradictions it caught more than 70% of errors in papers' main claims, about four times the best prior system. On real withdrawn papers and manipulated submissions, the gains shrink.Oct 13·7 min read

Research · AnalysisOpenAI's Navier–Stokes claim, explained for builders: the verifier is the storyOpenAI says 10,000 AI agents found a blowup in the Navier–Stokes equations and proved it in Lean. What is checked, what isn't, and what it means for agent builders.Oct 9·6 min read

Nothing in this format on this page — try another filter or the next page.