OpenAI Is Pissing Off a Bunch of Mathematicians—Again
“There’s a perception of mobster behavior” from leading AI companies, one mathematician tells WIRED as OpenAI prepares to release more than 100 new solutions to unsolved problems.
Mathematicians are growing increasingly frustrated with OpenAI's handling of their work, according to recent reports. At a meeting earlier this year, attendees expressed a mix of excitement and concern over the company's plans to release hundreds of solutions it generated using AI. The group requested that OpenAI publish papers to allow mathematicians to absorb and utilize the results, but these recommendations appear to have been ignored.
OpenAI has stated that it plans to release the results on GitHub, drawing on advice from the Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study.
However, several leading mathematicians feel that OpenAI has learned little from previous controversies surrounding AI-generated mathematical solutions. They believe the company is merely showcasing its models in a competitive landscape with rivals like Anthropic, both preparing for major initial public offerings. In a contentious incident in September, OpenAI deployed thousands of AI agents to solve a $1 million Millennium Prize problem, allegedly based on rumors.
Mathematician Tristan Buckmaster accused OpenAI of exploiting his collaboration with Anthropic, and during an attempt to negotiate credit, a researcher at OpenAI seemed to suggest excluding the Anthropic employee from any potential paper.
Many mathematicians are frustrated that OpenAI and Anthropic continue to release results through blog posts rather than scientific papers, making them harder to verify and often excluding prior work from other mathematicians. Some have even set up new tools to help the community use and evaluate machine-assisted proofs, but OpenAI has not shown interest in adopting these initiatives.
Critics point to a lack of progress in OpenAI's advisory group, which was formed to assess and communicate new results while upholding academic and professional standards. While the company claims to be working collaboratively with the math community, mathematicians feel that the power has been concentrated in the hands of AI companies.
Written by urgent.news from Wired Business's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
