OpenAI faces mathematician backlash over Navier-Stokes proof attribution
Mathematicians are clashing with OpenAI over the company's unverified solution to the Navier-Stokes problem. Twenty-five Fields medallists signed an open letter condemning rushed AI proofs that bypass proper attribution. The backlash follows NYU professor Tristan Buckmaster's allegations that OpenAI discouraged crediting a collaborator and potentially used his Codex prompts.
Artificial Intelligence··Evening
The Navier-Stokes Claim and Allegations
OpenAI claimed to have solved the Navier-Stokes problem using roughly 10,000 agents, tens of millions of dollars of compute, and 88 hours. The proof remains unverified. NYU professor Tristan Buckmaster accused OpenAI of pressuring him to exclude a collaborator working for Anthropic from being credited. Buckmaster also questioned whether his own Codex prompts influenced OpenAI's system. While the company categorically denied this, Buckmaster stated their denial should be treated with great skepticism.[1], [2]
Fields Medallists' Warning
The incident has deepened broader unease within the field. Twenty-five Fields Medal winners signed an open letter arguing that the rush to announce solutions leaves no time for proper writeups, isolating new methods, or citing relevant earlier work. They warned that this dynamic creates severe attribution and plagiarism issues. Furthermore, they argued that without willing mathematicians, AI-conceived ideas cannot fully come alive and the human transmission chain is lost. The letter follows June's Leiden Declaration.[2]
Structural Tensions in Frontier Math
The clash highlights structural friction between frontier AI labs and academic researchers. Reports indicate that labs can spend tens of millions of dollars on inference to beat original researchers to a proof, an environment that rewards secrecy. Mathematicians increasingly fear their use of tools like Codex might be fed back into new models. Amid escalating criticism from researchers, OpenAI recently withdrew its sponsorship of a mathematics event at CalTech.[2]
