ai-for-science
Insilico’s AI-designed IPF drug shifts six proteomic ageing clocks in a trial reanalysis News
A Nature Biotechnology reanalysis found six proteomic ageing clocks moved in the younger direction in treated IPF patients receiving Insilico’s AI-designed rentosertib, without proving rejuvenation in healthy people.
Claude designed protein binders that worked about half the time News
Anthropic gave Claude Mythos 5.1 open-source protein design tools and sent its output to outside labs, where nearly 50 percent of its designs across 12 targets bound successfully -- against the 10 to 15 percent hit rate typical of the field.
The case against using transformers for physics News
Caltech's Anima Anandkumar argues that simulating the physical world at industrial resolution implies hundreds of billions to a trillion tokens of context, putting it permanently out of reach for transformers -- and that neural operators, which learn maps between functions rather than sequences, are already outrunning supercomputers on weather, climate and fusion.
A record elliptic curve now lists Claude as a collaborator News
The canonical public record page for elliptic curve ranks has added a 2026 entry at rank 30 or higher, publishing an explicit curve with 30 independent points, and the attribution credits Claude alongside two named researchers, though no primary source describes what the model actually did.
De novo protein design: how a computer invents a molecule that sticks Lesson
De novo protein design means inventing a protein that does not exist in nature and that binds a chosen target, by predicting how a chain of amino acids will fold and then searching for a chain that folds into a shape which fits. It works because structure prediction got good enough to grade designs before anyone builds them.
Claude designed protein binders against 14 of 15 targets and two labs built every one of them News
Anthropic said on August 18 that Claude designed minibinders against 15 protein targets and succeeded on 14, with hit rates of 26.7 and 22.6 percent against the 10 to 15 percent typical of protein design campaigns, validated by two independent contract labs.
Bayesian optimization: how to search when every guess is expensive Lesson
Bayesian optimization is a strategy for finding the best setting of something when each trial is slow or costly: build a cheap statistical model of what you have learned so far, use it to predict both the value and the uncertainty of every untried option, then test whichever option best balances promise against ignorance.
AlphaEvolve helped tighten the matrix multiplication exponent, and the proof was checked in exact arithmetic News
A new paper establishes a certified upper bound of 2.371177 on the matrix multiplication exponent, improving the previous best of 2.371339, by reformulating the core optimization problem and refining the resulting algorithm with DeepMind's AlphaEvolve.
A discovery system that refuses to trust the language model's own confidence News
Large Discovery Models pair a generative proposer with a Bayesian surrogate that scores candidates from real experimental evidence rather than model confidence, reporting a 2.4 times greater reduction in a training metric and an 18.2 percent relative improvement in antibody binding energy.
Claude raised the zeta critical-line bound to 67.2 percent, and Anthropic published the proof News
An unreleased research version of Claude raised the proven lower bound on the fraction of Riemann zeta zeros lying on the critical line from 41.6 percent to 67.2 percent, and Anthropic published the paper and a machine-checked Lean proof on August 10.
An AI tightened a 70-year-old constant, and the paper says its judgment was the weak part News
A case study from seven researchers documents how an AI system helped tighten the best known bounds on the Grothendieck constant, and reports plainly that the system was strong at technical execution but weak at research judgment and at tracking where the work stood.
The White House science blueprint reroutes research toward AI - without moving a dollar yet News
The White House Office of Science and Technology Policy released Science: A New Golden Age on July 21, proposing to reorient federal research around individual investigators, fast grants, AI-ready data and automated labs - an agenda-setting document that specifies no reallocation of funds.
Arcee Announces Genesis-Science-1, an American Open-Weight Model for Science News
Arcee announced Genesis-Science-1, a planned trillion-parameter-class American open-weight scientific model built with the Department of Energy and national labs, with weights promised later this year.
AI Helped Crack a Famous Math Conjecture, and Humans Verified It in Lean News
Mathematicians found an explicit counterexample disproving the Jacobian conjecture in three dimensions, checked partly with an AI chatbot and formalized in a Lean proof, while two other viral AI-math claims remain unverified.
Isomorphic Labs Unveils a Drug-Design AI That Beats Physics at Predicting How Drugs Bind News
Isomorphic Labs, the DeepMind drug-discovery spinout, unveiled a Drug Design Engine that surpasses even physics-based methods at predicting binding strength and recomputed a cereblon drug pocket that took 15 years to find in the lab.
AI is now solving hard math and physics problems faster than humans can formally check them News
A widening 'verification lag' is emerging as AI produces candidate solutions to hard problems faster than experts can formally verify them - physicist Yuji Tachikawa reports Fable cracked a six-month research blocker, while a GPT-5.6 Erdos claim circulates without peer review.
AI supercharges individual science careers -- and quietly narrows what the whole field explores News
A Nature study of 41 million papers found that scientists who use AI publish far more and get far more citations, but AI-heavy research clusters around the same popular problems, shrinking the collective diversity of ideas the field pursues.
A 2025 Nobel chemist is leaving the US to run an AI materials lab in China News
Omar Yaghi, who won the 2025 Nobel Prize in Chemistry, has taken a full-time position at Tsinghua University in Beijing to lead a new AI-assisted materials-discovery institute, citing US grant cuts and a lack of American engagement with AI.
SciReasoner is an AI for science that shows which atoms and bonds its answer rests on News
Shanghai AI Lab's SciReasoner turns proteins, molecules, and crystals into discrete tokens the model reasons over out loud, so a scientist can audit which structural evidence its prediction depends on -- and expert reviewers rated its reasoning at least as good as a frontier LLM's in 98% of cases.
Anthropic launches Claude Science, an AI workbench that keeps data in the lab and checks its own citations News
Anthropic released Claude Science, a beta workbench that wires Claude into researchers' real tools -- PubMed, Jupyter, HPC clusters -- runs on the lab's own hardware so sensitive data never leaves, and pairs a working agent with a separate reviewer agent that flags and corrects citation and calculation errors.
SciReasoner, a science AI whose reasoning experts prefer 98% of the time News
SciReasoner, a multimodal scientific foundation model that turns molecular and material structures into a shared vocabulary, hit state-of-the-art on 67 of 86 benchmarks, and in blind review domain experts preferred its explanations over frontier LLMs in 98% of cases.
An AI Agent Screened 2.4 Million Crystals and Found Four New Superconductors News
An AI system called ElementsClaw screened 2.4 million candidate crystal structures and flagged four new materials that a lab has since synthesized and confirmed are genuine superconductors.
Can an AI Agent Reproduce Real Science? A New Test Says: Rarely News
A new benchmark points coding agents at the actual computational results behind ninety papers in top journals. The strongest models matched the published science on fewer than one in five.
Virtual Lab Tool
The open-source multi-agent research framework behind the Nature nanobody paper, where an LLM principal investigator coordinates specialist agents over tools like ESM, AlphaFold-Multimer and Rosetta. Runnable on your own project with your own agent roster.
Large Discovery Models Tool
Released code for a discovery loop that pairs a generative proposer with a Bayesian surrogate scored on real experimental results rather than model confidence, applied to molecules, antibodies and training configurations.
Apodex Discovery Tool
Executable environments built from real industry problems, with a rubric that scores an investigation's tools, repair, alternatives, coherence, evidence and scope independently of whether the final answer was right.