Mathematician alleges OpenAI may have used his private Codex sessions in claimed Navier-Stokes resolution

jm_alexia · x · 2026-09-08

Princeton mathematician Tristan Buckmaster is alleging that OpenAI's claimed resolution of Navier-Stokes may have drawn on his and Levent Alpöge's private Codex sessions. Per the account: he spent a year attacking the millennium problem via Codex, with all project drafts going through those sessions; a rumor then leaked that AI had solved it. After he emailed OpenAI, they called within days saying their model had also solved it — using his exact approach. When asked directly whether they trained on his sessions, OpenAI gave no answer. Unverified single-side allegation, but a major data-privacy flashpoint.

Related event: Mathematician accuses OpenAI of racing to publish Navier-Stokes result and demanding co-author removal(9 posts)→

Original post →

More from AGI Musings

AGI Musings channel →