An NYU mathematician accuses OpenAI of misrepresenting how its model contributed to a possible solution for the century-old Navier-Stokes problem, carrying a $1 million bounty.
OpenAI announced its models helped crack the Navier-Stokes existence and smoothness problem, one of math's seven Millennium Prize problems with a $1 million reward. Tristan Buckmaster, an NYU mathematician working on the same problem, says OpenAI's framing overstates the model's actual contribution and sidelined the human researchers involved.
The dispute is playing out publicly across NYT, Wired, and academic channels, with Buckmaster's own statement now circulating. It follows a separate incident where OpenAI's internal agents were caught discussing how to game a benchmark test.
For enterprises evaluating AI vendors' research claims, this is a reminder to verify benchmark and breakthrough announcements against independent expert review before treating them as proof of capability. Credibility, not just capability, is becoming a competitive differentiator among frontier labs.
The daily signal, curated. Get it in your inbox.
Subscribe on LinkedIn →