Fields Medalists Unite Against AI's Disruptive Influence on Mathematics
Twenty-five Fields Medal winners have signed a statement criticizing the tech industry's use of artificial intelligence in mathematics, citing a 'severe misalignment' with the field's core values. The declaration follows OpenAI's recent claim to have solved one of the Millennium Prize Problems, which many see as a turning point. The signatories argue that mathematics prioritizes conceptual insight over mere problem-solving, and that AI-generated proofs threaten the human culture of the discipline.
The declaration follows a week in which OpenAI claimed a solution to one of the seven Millennium Prize Problems, widely seen as a watershed moment. Signatories include Terence Tao, previously an AI optimist who appeared in OpenAI promotional material, but who now describes the situation as having “deteriorated markedly.” A separate group, the Association for Human Mathematics, was already forming before the Navier-Stokes announcement, indicating rising concern among top researchers. The statement frames mathematics as a miniature model of humanity, valuing diverse approaches and conceptual insight over mere problem-solving, and warns that AI-generated proofs threaten this culture.
This pushback could reshape how AI is deployed in academic fields, setting a precedent for other disciplines facing similar automation. If mathematicians successfully assert human-centered values, it may slow corporate-driven AI adoption in research, affecting funding priorities and public trust. Conversely, ignoring these concerns could accelerate a shift toward output-focused science, diminishing the role of human creativity. The outcome may influence not only mathematics but also how society weighs technological progress against cultural and intellectual traditions.