OpenAI Claims Solution to Millennium Prize Math Problem Amid Controversy

OpenAI has announced that it solved the Navier-Stokes problem, a major mathematical puzzle regarding fluid movement that has remained unsolved for nearly a century. The company utilized an internal system more powerful than its GPT-6 Astra model, deploying up to 10,000 AI agents that reached the solution in 88 hours. Mark Chen, head of research at OpenAI, stated that the computing power required for the endeavor cost millions of dollars. The solution was later verified by GPT-6 Astra in approximately 17 hours. The Navier-Stokes problem is one of seven Millennium Prize Problems, which carry a $1 million reward, though OpenAI stated it does not intend to claim the prize.

The announcement has been met with accusations of misconduct from mathematician Tristan Buckmaster of New York University and researcher Levent Alpöge of Anthropic. Buckmaster claims that OpenAI accelerated its efforts after learning that he and Alpöge were poised to announce their own breakthrough. He expressed concern that their work was stored in OpenAI's Codex model, potentially making it visible to the OpenAI team. Buckmaster further alleged that OpenAI offered a proposal where he could publish the solution using an internal OpenAI model but without Alpöge's name included.

OpenAI researcher Sebastien Bubeck denied that the company accessed the pair's work or inspected Codex prompts. However, a blog post from the company noted that it cannot rule out that de-identified data from the pair's usage of OpenAI products helped improve its models. Bubeck and CEO Sam Altman defended the team's work, and Bubeck specifically pushed back against the claim that the company suggested removing Alpöge's name. The breakthrough comes as AI continues to reshape mathematics, following similar claims of progress from Google DeepMind and previous OpenAI efforts.