When OpenAI announced progress on the Navier-Stokes Millennium Prize Problem — one of mathematics' most storied unsolved challenges — the response from leading mathematicians was not wonder but wariness. A dispute centered at New York University has laid bare a deeper tension: between the speed at which artificial intelligence can generate results and the deliberate, skeptical patience through which mathematical truth has always been established. The conflict is less about any single proof than about who gets to define what counts as knowing something, and whether the ancient standards of rigo