Mathematicians warn AI labs threaten open research culture in open letter

Twenty-five Fields Medalists signed an open letter arguing that AI labs are threatening the intellectual work of mathematicians by rushing unverified solutions to famous problems. The signatories warn that such solutions are often announced without adequate documentation, isolating new methods, failing to cite prior work, and leaving questions of attribution and plagiarism unresolved. They emphasize that without human mathematicians to develop and integrate AI-conceived ideas into the mathematical canon, the crucial transmission chain between researchers could be lost.

The letter follows accusations by NYU professor Tristan Buckmaster, who claimed OpenAI pressured him not to credit an Anthropic collaborator for solving an important math problem. Buckmaster also suggested OpenAI may have used its Codex tool to produce its own groundbreaking proof during a marathon weekend of inference. OpenAI denied the allegations, but the incident underscores broader concerns about transparency and the use of AI tools in mathematical research.

OpenAI escalated the dispute by withdrawing its sponsorship of a math event at CalTech after researchers at the university criticized the company. The move reflects the growing friction between AI labs and the mathematical community over the role of AI in research and the standards for verifying and crediting new proofs.

The open letter and related controversies come amid broader debates about how AI is changing scientific and creative professions. The signatories argue that the rush to leverage AI for breakthroughs risks undermining the open research culture that has long underpinned progress in mathematics and other fields. They warn that if AI continues to reshape research workflows without safeguards, the fundamental goals of scientific and creative work—such as transparency, collaboration, and the transmission of knowledge—could be compromised.