OpenAI’s Navier-Stokes claim puts AI research credit under pressure

OpenAI says an internal, unreleased AI system has produced a proof for the Navier-Stokes problem, a famously stubborn question about whether the equations used to model air and water can break down even from smooth conditions. The company released a written argument and a version checked in Lean, software that verifies each logical step, but the result is still a claim: Clay Mathematics Institute rules require publication, two years, and broad acceptance before any Millennium Prize is awarded. The announcement collided with work by NYU mathematician Tristan Buckmaster and Anthropic researcher Levent Alpöge, who say OpenAI learned of their unpublished progress and then raced to claim the broader solution; OpenAI denies seeing their work and says its proof took a fundamentally different route. The immediate question is therefore not only whether AI can reach the frontier of mathematics, but whether academics can safely use frontier-lab tools on private research without losing control of credit.