The Association for Human Mathematics is calling on mathematicians to boycott OpenAI following the company’s reported release of more than 700 mathematical research files. The organization argues that the move raises serious questions about how advanced mathematical research is being handled and what role artificial intelligence companies should play in the development and distribution of mathematical knowledge.
According to the association, mathematicians should reconsider their relationships with OpenAI and avoid supporting the company’s research ecosystem until concerns surrounding the release are addressed. The boycott appeal puts renewed attention on the increasingly complicated relationship between academic researchers and AI companies developing systems capable of working with advanced mathematics.
The reported release of more than 700 files is significant because mathematical research is often the product of years of work involving researchers, institutions and specialized communities. Research files can contain proofs, experimental results, technical notes, datasets, code and other material that may be valuable to researchers working on related problems. Questions over how such material is obtained, processed or released can therefore have consequences beyond the individual documents involved.
The Association for Human Mathematics appears to be positioning the issue as part of a broader debate about the treatment of human mathematical work in the age of artificial intelligence. Rather than focusing solely on the technical capabilities of AI systems, the boycott demand shifts attention toward the people producing the underlying research and the standards that should govern the use of their work.
OpenAI Faces Growing Questions Over AI and Mathematical Research
The dispute comes at a time when AI companies are investing heavily in mathematical reasoning. Advanced AI models are increasingly being tested on difficult problems involving formal proofs, mathematical reasoning, coding and scientific research. OpenAI has been among the companies attempting to push these capabilities forward, making mathematics an important part of the broader AI race.
For mathematicians, however, the rapid development of AI creates a difficult question: how should researchers’ work be used by increasingly powerful models? The issue becomes particularly sensitive when research material is made available at scale, since mathematicians may want clearer information about attribution, licensing, consent and the eventual use of their work.
The Association for Human Mathematics’ boycott call therefore represents more than criticism of a single release. It reflects a wider concern about whether traditional academic expectations are keeping pace with AI development. Mathematical research has historically relied heavily on open communication and the sharing of ideas, but researchers may also expect appropriate recognition and control over how their work is reproduced or incorporated into commercial technologies.
The reported release of 700-plus mathematical research files could intensify that debate if more researchers begin questioning how AI companies source and distribute mathematical material. At the same time, the specific details surrounding the files, including their origin, licensing status and the circumstances of their release, will be important in determining the strength of the association’s objections.
Related: Trump Secures Voluntary AI Safety Accord With Meta, Anthropic, Nvidia and OpenAI CEOs
A boycott could also have practical consequences for AI research. Mathematics researchers contribute to academic collaborations, benchmark development, model evaluation and the creation of datasets used to test AI reasoning. A sustained withdrawal of participation would potentially make it more difficult for AI companies to maintain relationships with parts of the mathematical research community.
For OpenAI, the controversy adds another layer to an already significant discussion surrounding AI and intellectual property. The company and other AI developers have faced broader scrutiny over the use of human-created material to train and evaluate artificial intelligence systems. Mathematical research presents a particularly interesting case because much of it is publicly published, while the underlying work, notes and research artifacts can involve substantially more material than what appears in a final academic paper.
Whether the boycott gains meaningful support among mathematicians remains to be seen. For now, the Association for Human Mathematics’ statement highlights a growing fault line between the rapid expansion of AI capabilities and the expectations of researchers whose work contributes to the knowledge those systems are increasingly designed to understand.
The controversy also raises a larger question for the AI industry: as artificial intelligence becomes better at producing and manipulating mathematical knowledge, what obligations should companies have toward the human researchers whose work forms part of that knowledge base? The answer could shape how future AI research collaborations are structured, particularly in fields where years of specialized human expertise can become valuable training and evaluation material for increasingly capable models.
About The Author
