OpenAI Claims Navier-Stokes Proof Amid Credit Fight — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- OpenAI announced an AI-generated solution to the Navier-Stokes equations, one of the unsolved Clay Millennium Prize problems, each worth $1 million.
- Sebastien Bubeck said OpenAI began training a new model on August 28 after hearing rumors of Anthropic's progress, then deployed over 1,000 agents for more than 50 hours before reaching a Lean-formalized solution on Sunday.
- Tristan Buckmaster (NYU) and Levent Alpöge (an Anthropic researcher) posted their own advances Monday in a related area, saying they used Claude and Codex to complete the work.
- Buckmaster alleged OpenAI learned of their progress last week and offered proposals, including one to publish a paper crediting the breakthrough to an internal OpenAI model without naming Alpöge.
- Buckmaster also said he asked whether OpenAI had accessed his and Alpöge's Codex logs and was told the model "didn't look up user data," but received no answer about training.
- OpenAI denied inspecting the pair's Codex prompts or using their work, with Bubeck calling the competing achievement "monumental" and Ven Chandrasekaran saying the company's solution was "significantly different in nature."
- Mark Chen, OpenAI's head of research, said the compute required to solve the problem cost "in the millions of dollars."
Why it matters: This is the first high-stakes credit dispute over an AI-generated proof at the Millennium Prize level — a domain where authorship carries a $1 million prize and lasting academic recognition. Buckmaster's allegation that OpenAI sought to publish without crediting Alpöge raises a new question: when an AI lab's model and a human researcher race toward the same theorem, who owns the answer, and who decides whose name appears on the paper?
Ask SkimNews
