Top mathematicians to advise OpenAI on AI maths results
OpenAI says a new internal model has resolved more than 100 long-standing open problems in mathematics. Nine mathematicians, including Timothy Gowers and Edward Witten, have formed an independent group to advise it on releasing the results. The group states it has no decision-making power.
September 23, 2026 - 3:00 pm
Credit: Evolf via Shutterstock.com
"We are currently facing the very specific challenge of advising OpenAI on how to coordinate the release of a large number of significant results in mathematics that they report have been produced by their internal model," the Advisory Group on Mathematics and Artificial Intelligence wrote in its launch statement.
OpenAI Announces Advisory Group
OpenAI announced the formation of this group on Monday. The model, trained starting August 28th, reportedly resolved over 100 long-standing open problems across most areas of mathematics, in addition to the Navier-Stokes result announced earlier this month.
According to OpenAI, the progress surprised their own mathematicians, leading to internal discussions on how to inform the community and prepare for these developments.
The Advisory Group
The group is hosted at the Institute for Advanced Study in Princeton and is available online at agmai.org. Its initial members include:
- François Charles (ENS-PSL)
- Camillo De Lellis (IAS, GSSI)
- Timothy Gowers (Collège de France, Cambridge)
- Martin Hairer (EPFL, Imperial College London)
- Nikhil Srivastava (Berkeley, Simons Institute)
- Ulrike Tillmann (Oxford, INI)
- Ravi Vakil (Stanford)
- Edward Witten (IAS)
- Melanie Matchett Wood (Harvard)
OpenAI approached some members about establishing an external advisory board, but they opted to form an independent group instead, inviting others to join.
Group Responsibilities and Limits
The group will assist OpenAI in assessing the significance of new results, advising on their release, and considering academic and professional standards. It can also offer unsolicited advice, comment on OpenAI’s impact on mathematics, and publish its recommendations.
However, the group emphasizes that it "will not be responsible for advising [OpenAI] on how to pace our internal progress on mathematics" and has no decision-making power over any AI company.
Background and Concerns
OpenAI cites a recent open letter, "A Severe Misalignment of AI in Mathematics", which raised concerns about using the solution of open problems as a benchmark for AI systems. Twenty-five Fields Medal winners signed this letter, according to TechCrunch, though only Camillo De Lellis is part of the new group.
Researchers expressed mixed sentiments towards the group, with many noting they still had questions regarding its scope and OpenAI’s responsiveness to its advice.