PRINCETON, NEW JERSEY / RankWire.AI / – OpenAI has initiated a partnership with a standalone group of mathematicians to oversee the review and dissemination of results generated by AI in the field of mathematics. The Advisory Group on Mathematics and Artificial Intelligence is based at the Institute for Advanced Study in Princeton. Its role includes advising on new findings, academic standards, and the application of OpenAI tools in research and education. The group functions independently from OpenAI, with its members not receiving compensation from the organization.
The advisory panel is empowered to offer guidance outside of OpenAI’s requests and can publish its suggestions publicly. It also has the capacity to comment openly on how the company’s work influences the field of mathematics. Membership decisions are controlled by the group itself, which may also advise other AI firms whose models have a substantial impact on mathematics. Notably, the group holds no decision-making authority at OpenAI or any other AI entity. Furthermore, OpenAI has clarified that the group will not provide advice regarding the pace of its internal mathematical research efforts.
The initial team consists of nine mathematicians: François Charles, Camillo De Lellis, Timothy Gowers, Martin Hairer, Nikhil Srivastava, Ulrike Tillmann, Ravi Vakil, Edward Witten, and Melanie Matchett Wood. Their affiliated institutions include the Institute for Advanced Study, Harvard, Stanford, Oxford, Berkeley, Cambridge, EPFL, and ENS-PSL. De Lellis and Witten serve as principal investigators for the project at the Institute for Advanced Study that hosts this initiative.
OpenAI reports swift advancements in mathematical research
OpenAI disclosed the formation of this advisory body amid new claims related to an internal artificial intelligence model. The company announced that training on this system commenced on August 28. Since then, OpenAI asserts that the AI has solved over 100 longstanding open problems across a broad spectrum of mathematics. The company describes these accomplishments as the output of its internal system. The advisory group notes that its immediate task involves guiding how OpenAI should coordinate the disclosure of these significant results.
The announcement follows OpenAI’s release on September 8 of research related to the Navier-Stokes existence and smoothness problem, one of the Millennium Prize Problems in mathematics. OpenAI provided an analytical proof and formalization using the Lean proof system. The company claimed that its internal model generated the proof, demonstrating that a smooth fluid could develop a singularity within finite time. OpenAI also clarified that it does not plan to claim the Millennium Prize for this achievement.
The independent panel emphasizes standards for review and publication
Following discussions with some of its members about forming an external advisory board, OpenAI led the mathematicians to establish a fully independent group instead. This panel has since expanded by inviting additional members and aims to facilitate collaboration between AI organizations, the mathematics research community, and broader scientific circles. The group’s focus includes promoting responsible approaches to the presentation and dissemination of mathematical outcomes. Its formation is part of a wider conversation among mathematicians on how AI-generated research should be evaluated, credited, and communicated.
Currently, the group’s work centers on the mathematical results claimed by OpenAI’s internal system. It seeks input from mathematicians to shape guidance on how to responsibly release these findings. The group commits to publishing its recommendations on its website, while responses from mathematicians will remain confidential unless they approve their publication. Final decisions will stay with the companies that receive the advice, with the independent panel acting as a liaison between the mathematics community and the AI industry.