Mathematicians form an advisory group as OpenAI claims 100 solved problems

Nine mathematicians have formed an independent Advisory Group on Mathematics and Artificial Intelligence, hosted at Princeton's Institute for Advanced Study and online at agmai.org. Terence Tao, announcing the group on his blog, describes its job plainly: to advise AI companies on how they work with mathematical research and the wider mathematical community, and on how they present mathematical results responsibly. The members include Timothy Gowers, Martin Hairer, Edward Witten, Ravi Vakil, Camillo De Lellis, and Melanie Matchett Wood. They take no payment, stay independent of any AI company, and will publish their recommendations in public.

The timing is not an accident. OpenAI says its internal model has resolved more than 100 open problems in mathematics, and has claimed a solution to the Navier-Stokes Millennium Prize problem. The advisory group's first concrete task is to help OpenAI decide how to release such a large batch of claimed results in a way the field can actually check. This lands amid real tension: 25 Fields Medalists recently signed an open letter warning that the race to solve famous problems could damage mathematical scholarship. Only one member of the new group, Camillo De Lellis, also signed that letter.

Tao and colleagues are asking mathematicians to send feedback through a form on the group's site. The group has no say over how fast OpenAI moves internally, a limit several commenters flagged as the point where its influence runs out.

Why it matters

If you work in or near mathematics, a flood of machine-produced proofs is coming, and peer review was never built for that volume or that source. This group is the first serious attempt to set norms for checking and publishing such results, so how it handles OpenAI's claims will shape whether you can trust the next "AI solved X" headline.