OpenAI forms mathematics advisory board after claiming Navier-Stokes solution

On Monday OpenAI announced the Advisory Group on Mathematics and Artificial Intelligence, an independent body that will sit at the Institute for Advanced Study in Princeton, New Jersey. The group is meant to give mathematicians a voice in the company's decision-making around math-focused research and to act as a bridge between the scientific community and the broader public.

The move follows OpenAI's sudden publication of a solution to the Navier-Stokes problem, one of the seven Millennium Problems for which the Clay Mathematics Institute offers a million-dollar prize. Beyond that, the company says the same internal model has solved more than 100 additional open problems across most areas of mathematics. OpenAI has released no benchmarks or details of the problems, leaving the claims as unverified vendor claims.

Earlier this month 25 Fields Medal winners signed an open letter accusing AI labs of threatening their intellectual work by racing to solve famous problems at a breakneck pace. The letter reflects a deep unease: the rush to publish first is sidelining the verification, replication and discussion that are standard in mathematics.

According to OpenAI's post, the unpaid members may evaluate the significance of new results, coordinate their publication, offer unsolicited advice, speak publicly and manage their own membership. The company was explicit, however, that the group will not advise on the pace of OpenAI's internal progress in mathematics. The Institute for Advanced Study underscored in its own statement that it has no decision-making authority over any AI company; responsibility remains with the companies themselves.

Nine prominent mathematicians were named as initial members. Only one, Camillo De Lellis of the Princeton institute, also signed the Fields Medalists' letter. The fact that most members did not join the public protest raises questions about the group's practical independence, given that its powers are limited from the outset to advisory roles with no ability to halt or steer research.