Artificial Intelligence
Mathematicians call for OpenAI boycott over AI proofs
The Association for Human Mathematics urged researchers to shun OpenAI after the company released hundreds of AI-generated manuscripts, prompting dispute over verification, authorship and the role of AI in mathematics.
- The Association for Human Mathematics called for a boycott of OpenAI after the company published hundreds of AI-generated mathematical manuscripts.
- Terence Tao, who chairs the group, shared the statement and described a "Math 2.0" era.
- OpenAI had earlier said an internal model solved more than 100 open mathematics problems in one month, and it later set up AGMAI at the Institute for Advanced Study.
- The article said unnamed sources reported AI companies were probing cryptographic weaknesses, with about 8,000 problems tested and a success rate of roughly five per cent.
The Association for Human Mathematics said OpenAI had ignored advice that advanced problems should not be tried on internal models. It said the release of more than 700 files was a display of power rather than scholarship.
Tao shared the statement on his blog and linked it to a "Math 2.0" view on Mastodon, while Scott Aaronson called the situation a "Mathocalypse" on his blog. OpenAI had earlier said its internal model had solved more than 100 open problems in one month, including work on the Navier-Stokes Millennium Problem.
After criticism, OpenAI created AGMAI at the Institute for Advanced Study, but the group had no control over the company's research pace. The AHM statement also raised copyright, attribution and plagiarism concerns, and Aaronson contrasted OpenAI's approach with Anthropic's collaboration with two algorithm researchers.
Aaronson also cited his wife, Dana Moshkovitz, on a claimed proof of the Unique Games Conjecture, which she said was hard to follow and used a strange construction. He added, citing unnamed sources, that AI companies were probing cryptographic protocols and primitives, and the article said the model involved was likely OpenAI's current internal one.
Named in this story
People
- Terence Tao
- chairs the Association for Human Mathematics and shared its statement
- Scott Aaronson
- blogged about the dispute and compared publication models
- Dana Moshkovitz
- criticised the claimed proof of the Unique Games Conjecture
- Peter Scholze
- signed an earlier warning statement
- Maryna Viazovska
- signed an earlier warning statement
- Martin Hairer
- signed an earlier warning statement
Companies
- OpenAI
- published hundreds of AI-generated mathematical manuscripts
- Anthropic
- worked with two researchers on an alternative proof process
Organisations
- Association for Human Mathematics
- called for a boycott and accused OpenAI of breaking research norms
- AGMAI
- OpenAI's advisory group at the Institute for Advanced Study
- Institute for Advanced Study
- hosts AGMAI
How the source tells it
The piece was combative and cautionary, with vivid metaphors, quoted judgments and a few unattributed claims pushing it beyond plain reporting.
- alarm apocalyptic labels and disaster metaphors are used to frame the effect on mathematics
- moral framing the story leans on accusations about research norms, attribution and plagiarism
- speculation presented as fact claims based on unnamed sources are presented as part of the main account