AI breakthroughs shake mathematics, sparking existential debate among leading scholars

OpenAI recently publicized a set of results from an internal model, referred to as Astra, that it says resolves ten major open problems in mathematics and theoretical computer science. The claim generated a storm of reaction: many senior mathematicians, including Fields Medalist James Maynard, expressed surprise and concern about the implications for the profession, while others praised the technical achievement. Critics pointed to vague attribution in the accompanying paper and questioned whether the breakthroughs were truly original or built on existing work.

The episode has ignited a broader discussion about whether AI will simply become a powerful tool or fundamentally reshape the purpose and funding of mathematical research. Some scholars note that AI’s current weaknesses in basic arithmetic and certain subfields, such as topology, suggest a uneven impact. The debate also touches on issues of access, cost, and the potential for AI-driven research to bypass traditional academic pathways.