25 Fields Medalists Sign Joint Letter Warning of AI Race's Impact on Mathematical Research Ecosystem

Twenty-five Fields Medalists recently signed an open letter warning that AI companies’ race to solve famous mathematical problems could damage mathematical research and its knowledge ecosystem. The letter points out that the value of mathematical work lies not only in reaching correct conclusions but also in refining methods, explaining ideas, citing prior achievements, and integrating knowledge into the discipline through teaching and peer exchange; hastily publishing AI-generated proofs may trigger issues regarding authorship, plagiarism, and review responsibility. The controversy intensified after OpenAI announced a proof of the Navier-Stokes Millennium Prize Problem. New York University professor Tristan Buckmaster alleged that OpenAI had pressured him not to list co-researcher Levent Alpöge, an Anthropic employee, as a contributor to the result, and questioned whether his own Codex usage records influenced OpenAI’s research. OpenAI denied this possibility, stating that an internal investigation confirmed relevant prompts were not used for this outcome, noting that its system consumed approximately 130 billion output tokens to complete the proof, while formal conclusions still await further scrutiny by the mathematics community. Subsequently, OpenAI withdrew sponsorship for a Caltech math event. The signatories argue that if AI laboratories secure results first thanks to massive computing power, researchers might turn toward secrecy, thereby undermining the long-established culture of open collaboration within the mathematical community. They call on enterprises, academic institutions, and policymakers to improve mechanisms for attribution, transparent disclosure, and independent verification, ensuring that AI promotes understanding in mathematics rather than merely pursuing faster answers.