ChatGPT solves 377 math problems; Mathematicians call for boycott
Mathematicians are calling for a boycott of OpenAI after the company released hundreds of AI-generated mathematical findings and more than 700 files, triggering a backlash over its approach to scientific research.
The Association for Human Mathematics has urged researchers to stop collaborating with OpenAI, accusing the company of pursuing research that the mathematical community never requested and undermining the credibility of scientific work.
The controversy follows OpenAI’s claim last month that its internal model had solved the Navier-Stokes Millennium Prize Problem, one of mathematics’ most complex and long-standing unsolved challenges. The company subsequently released findings relating to 377 mathematical problems, along with more than 700 files.
The association criticised the scale and manner of the release, arguing that publishing such a large volume of material demonstrated technological power rather than genuine academic excellence. It also accused OpenAI of disregarding recommendations from an independent advisory committee hosted by the Institute for Advanced Study in Princeton.
A major concern is that AI systems are producing mathematical arguments that human researchers may struggle to understand, verify for accuracy or take responsibility for. The association also warned that unequal access to advanced AI systems and computing resources could deepen divisions within the mathematical community, giving researchers at major technology companies an advantage over others.
OpenAI has defended its approach, stating that it discussed the findings with the advisory committee and followed its recommendations before publication. The company said the results were published through GitHub and used Lean, a programming language that enables computer-assisted verification of mathematical proofs.
OpenAI also indicated that producing an average result required computing resources equivalent to three hours of reasoning by ChatGPT Pro. However, the dispute has intensified concerns about transparency, accountability and fairness in AI-driven mathematical research.