OpenAI reportedly asked a mathematician to drop his Anthropic coauthor to publish a model-made proof

rickasaurus · x · 2026-09-08

A strange tale in the OpenAI-Anthropic feud, as relayed by Hesamation: NYU mathematician Tristan Buckmaster and a coauthor (who works at Anthropic) spent a month with AI assistance proving a major Euler blowup result. After rumors spread that Anthropic had solved a major open math problem, Buckmaster found his project had reached OpenAI and contacted them. Days later, OpenAI said their model had produced a 100-page Navier-Stokes proof and asked him to publish it—on the condition he drop his Anthropic-employed coauthor. He refused; OpenAI allegedly replied, "why would you ruin your career?" The poster calls it a gross look for OpenAI. Unverified anecdote.

Related event: Mathematician Claims LLM-Assisted Breakthrough, Accuses OpenAI of Scooping and Co-Author Removal(8 posts)→

Original post →

More from Companies & People

Companies & People channel →