NYU mathematician accuses OpenAI of dirty play on Navier-Stokes proof
Original: OpenAI fought dirty on career-making math problem, says NYU mathematician
Why This Matters
The dispute raises serious questions about AI lab conduct and research integrity in high-stakes mathematics.
NYU professor Tristan Buckmaster announced three math proofs Tuesday, including a preliminary result on the Navier-Stokes Millennium Prize problem, while alleging OpenAI used leaked information about his approach to race ahead with massive compute resources to claim a full proof first.
NYU mathematics professor Tristan Buckmaster, collaborating with Anthropic mathematician Levent Alpöge and using both OpenAI's Codex and Anthropic's Claude, announced three proofs on September 8, 2026, including a preliminary finding on the Navier-Stokes existence and smoothness problem — one of seven Clay Mathematics Institute Millennium Prize problems, each carrying a $1 million bounty.
Buckmaster alleged that while he and Alpöge were finalizing their results, information about their progress was passed to OpenAI. When they inquired, OpenAI claimed it had already achieved a full proof. However, follow-up questions about the timing and degree of human input revealed evasive answers. Buckmaster says it eventually emerged that an entire OpenAI team had worked on the problem using 'an insane amount of compute,' and that their first prompt had been sent only days earlier — after details of Buckmaster's approach had reportedly reached the company.
The allegation implies OpenAI leveraged leaked knowledge of Buckmaster's unconventional approach (via the 'smooth force' route in Fefferman's problem statement) and its superior compute to reach a formal proof first. Sebastian Bubeck, who leads OpenAI's mathematical research, denied the claims, calling them 'false and inflammatory' and stating he engaged 'following academic norms.' Bubeck said a fuller statement would follow.