Fields Medalists Accuse OpenAI of Undermining Math Research

Twenty-five Fields Medal recipients have signed an open letter warning that artificial intelligence labs are threatening the integrity of mathematical research in their race to solve famous problems.

The letter comes amid escalating tensions between OpenAI and the mathematics community. NYU professor Tristan Buckmaster this week accused OpenAI of pressuring him not to credit a collaborator from Anthropic for solving an important math problem. Buckmaster also questioned whether OpenAI had used their work with its Codex tool to produce its own proof over a marathon weekend of inference. OpenAI subsequently withdrew its sponsorship of a mathematics event at CalTech following criticism from university researchers, according to TechCrunch.

The attribution crisis

The Fields medalists argue that while AI-generated solutions to mathematical challenges could benefit humanity, that potential depends on whether the broader math community can understand, verify, and communicate those solutions.

"Often these solutions are announced in a rush, leaving no time for a proper writeup, the isolation of new methods and ideas, and citing relevant previous work of others," the mathematicians wrote. OpenAI's contested proof remains unverified.

The letter raises concerns about attribution and plagiarism that mirror challenges facing creative professions. Without mathematicians to develop and integrate AI-conceived ideas into the mathematical canon, "AI-conceived ideas would never become fully alive and the crucial human transmission chain between mathematicians would be lost."

A chilling effect on collaboration

Mathematicians are growing concerned that their use of Codex may have been fed into OpenAI's newer models. This fear is driving a shift away from the field's traditionally open research culture.

The competitive dynamics are stark: frontier AI labs can now spend tens of millions of dollars using large language models to beat original researchers to a proof. This capability creates incentives for secrecy that run counter to how mathematical research has historically advanced.

The September letter follows the Leiden Declaration, released by a working group of mathematicians in June, which offered recommendations for mathematicians, institutions, and policymakers grappling with how LLM-generated proofs will reshape their work.

Why it matters

The mathematicians' concerns extend beyond their own field. They argue that the value of mathematical work lies not just in the proofs themselves, but in the intellectual infrastructure that trains students, generates new questions, and integrates discoveries into broader human knowledge. As AI tools transform workflows across disciplines—from software engineering to scientific research—the tension between speed and understanding, between computational power and human insight, will only intensify. The math community's struggle today previews challenges that other fields will face as AI capabilities advance.

"The issues the mathematical community faces now are similar to issues that other scientific and creative professions are facing, and indicate issues that all of humanity might face," the mathematicians wrote.

The details were first reported by Tim Fernholz at TechCrunch.