Tag
Briefings and ideas about AI scientific discovery from AI podcast discussions and the research they cite, with links to the original sources.
Biohub's Alex Rives expects pretraining, elicitation and RLVR to carry over from language models to biology, but says verification must happen in the lab. He is unsure whether biology and language models will share one latent space.
Biohub science head Alex Rives argues that no lab animal can ever be a perfect stand-in for people. He says an AI model trained across many species, as protein models were trained across evolution, could predict human biology far better.
Biohub's head of science Alex Rives argues that cell models are held back by what biology can measure. His examples are whole-zebrafish imaging and a laser phase plate for electron microscopes, so far tested only on isolated proteins.
Biohub's head of science, Alex Rives, says virtual cell models fall short above all because biology can't yet measure what matters. He expects an information-level cell model within about five years.
XPRIZE CEO Anousheh Ansari said a prize to store donor organs for months, tested in pigs with a six-month target, won this year's Visioneering. She and Peter Diamandis also described a funded telescope-data prize and less advanced ideas.
A preprint reports GPT-6 Astra got 49 of 50 unpublished physics calculations right with web access, for about $2 each. On Moonshots, XPRIZE CEO Anousheh Ansari asked who will find new problems if AI solves the defined ones.
Alexander Wissner-Gross forecast that virtual-cell AI will propose small-molecule cures for the top 5,000 diseases within two to three years. He said regulatory approval would be a harder, policy-dependent question.
OpenAI posted hundreds of AI-produced math manuscripts on October 6 and withdrew three a day later. A mathematicians' group objected; on Moonshots, Alex Wissner-Gross argued math is just the first field AI will bulk-solve.
Ian Quigley says co-folding models still place a molecule where a pocket was digitally destroyed, so Leash Bio engineers protein mutants for negative data. In a 20-protein exploratory analysis, its model's attention spread more evenly after retraining on mutants plus about 40% more ordinary data.
Ian Quigley of Leash Bio says AI agents querying its premeasured binding screens can search for strong binders within 100 picks, and argues this resists memorization better than scaling its Hermes model. Leash released a version, Minos, October 14.
GovAI's Alan Chan floated certified AI auditors that inspect a company or government, publish a redacted report, then have their memory wiped. He doubts they would be transformational, since training, hardware and sensors must still be trusted.
Eric Weinstein says AI should be free to challenge math and physics leaders he accuses of holding their fields back. He calls agent swarms on the internet dangerous. He believes a rigorously air-gapped, problem-solving AI would be safe to run, but he acknowledges that its research might give it new means of escape.