AI
OpenAI launches math advisory board amid backlash over rapid problem-solving
The company forms an independent group of mathematicians to advise on AI research at Princeton's Institute for Advanced Study, even as elite mathematicians object to the pace of its breakthroughs.
TechCrunch — AI
OpenAI has established a new advisory group aimed at bridging its artificial intelligence research with the mathematics community. The Advisory Group on Mathematics and Artificial Intelligence will operate independently through Princeton's Institute for Advanced Study and give academic mathematicians a formal channel to shape the company's mathematical research direction.
Why now?#
The timing reflects mounting tension over OpenAI's achievements. The company recently unveiled a solution to the Navier-Stokes Millennium Prize problem, one of the most coveted unsolved questions in mathematics. The announcement also revealed that the underlying AI model has cracked over 100 additional open problems spanning multiple mathematical fields. This flood of results has alarmed prominent mathematicians, who view the rapid-fire solutions as destabilizing to their discipline. Earlier this month, twenty-five Fields Medal recipients—mathematics' highest honor—issued an open letter warning that competitive AI labs are undermining traditional mathematical research by rushing to claim credit for foundational breakthroughs.
What the group will (and won't) do#
The new advisory body operates with built-in limitations on its influence. Members can offer guidance on the importance of new mathematical results and coordinate how discoveries are publicly shared, but they hold no authority over OpenAI's research pace or priorities. The company explicitly states that "the group will not be responsible for advising us on how to pace our internal progress on mathematics." The Institute for Advanced Study reinforced this constraint in its own statement, noting that while advisors can contribute perspective, "we do not have decision making power at any AI company."
Still, the structure grants members unusual autonomy for a corporate advisory role. Mathematicians selected for the group receive no compensation, can publicize their own views without restriction, and control their own participation—elements intended to preserve their independence from the company.
The elephant in the room#
Only one of the nine inaugural members—Camillo de Lellis from the Institute for Advanced Study—also signed the Fields Medalists' letter objecting to OpenAI's methods. This disconnect underscores the divide within the mathematical establishment over how to respond to AI disruption in their field.
