Ted Hisokawa
Sep 21, 2026 20:58
OpenAI partners with an independent advisory group to address concerns about AI’s role in solving major math problems and its broader implications.
OpenAI announced on September 21, 2026, the formation of an independent Advisory Group on Mathematics and Artificial Intelligence. The group includes prominent mathematicians and aims to guide the responsible review and communication of significant mathematical results generated by OpenAI’s internal AI models.
The announcement comes as OpenAI’s internal model reportedly solved over 100 longstanding mathematical problems, including the Navier–Stokes Millennium Prize problem. This pace of discovery has sparked serious discussions within the mathematical community about the implications of AI-driven problem-solving and how to adapt to these rapid advancements.
The Advisory Group, hosted by the Institute for Advanced Study, includes luminaries such as Timothy Gowers (Collège de France, Cambridge) and Edward Witten (IAS), alongside seven other renowned mathematicians. OpenAI clarified that the group operates independently, with full autonomy to critique the company’s work and publish their findings publicly. Notably, OpenAI will not pay members, reinforcing the group’s impartiality.
Criticism from the mathematical community has intensified in recent months, highlighted by the Leiden Declaration released in June 2026. Signed by the International Mathematical Union, the declaration urges AI companies to consider the long-term impact of their systems on mathematical research values and norms. The Advisory Group aims to address these concerns by acting as a bridge between AI researchers and mathematicians, ensuring transparency and responsible deployment of AI capabilities.
One of the group’s primary tasks is advising OpenAI on how to coordinate the release of its mathematical discoveries. OpenAI has emphasized its commitment to responsible development, recognizing that mathematics underpins a wide range of disciplines and applications. The group will also explore how AI tools can enhance mathematical research and education.
While the Advisory Group’s role is limited to advising and does not extend to influencing OpenAI’s internal research pace, its creation signals a growing recognition of the need for governance in AI-driven mathematics. The move also reflects broader industry concerns about balancing innovation with ethical considerations, particularly as AI systems increasingly tackle foundational scientific problems.
The mathematical community is watching closely to see how OpenAI and the Advisory Group navigate these challenges. As AI continues to reshape the boundaries of research, this collaboration could set a precedent for responsible AI integration across other scientific fields.
Image source: Shutterstock
Be the first to comment