OpenAI Forms a Mathematics Advisory Group of Nine Leading Mathematicians

OpenAI formed an independent Advisory Group on Mathematics and Artificial Intelligence, made up of nine leading mathematicians, to guide how the company evaluates and shares AI-generated mathematical results.

Details

• Members: François Charton, Camillo De Lellis, Timothy Gowers, Martin Hairer, Nikhil Srivastava, Ulrike Tillmann, Ravi Vakil, Edward Witten, and Melanie Matchett Wood
• What prompted it: an open letter titled “A Severe Misalignment of AI in Mathematics” raised concerns from mathematicians about using open problem-solving as an AI benchmark; the group was formed as a mechanism for dialogue between OpenAI and the mathematics community
• What the group does: evaluates emerging mathematical results and advises on their significance, coordinates how findings are shared publicly, ensures adherence to academic standards, and recommends ways AI tools can support mathematical research and education
• Independence: members exercise their own judgment, can offer unsolicited advice, and may publicly comment on OpenAI’s mathematical progress; they are unpaid and don’t direct the company’s internal research pace
• Context: OpenAI says its models have resolved over 100 longstanding mathematical problems since August 28, including a Millennium Prize problem — the Navier-Stokes result disclosed earlier in September

What happened next

The advisory group gives outside mathematicians a formal, ongoing role in vetting and contextualizing OpenAI’s math claims, addressing a credibility gap that had grown alongside the pace of the company’s mathematical announcements — including the Navier-Stokes disclosure, which OpenAI itself was careful to frame as not claiming the Millennium Prize.