OpenAI math turned into a live argument on September 8, when OpenAI said its models had solved one of math’s most famous problems. The announcement hit mathematicians as more than a technical milestone. It also forced a blunt question about who gets to count as a theorem-proving mind.
Scott Aaronson, a professor at the University of Texas, Austin, wrote that, “In whatever years I have left, I don’t expect that I’ll ever again prove a theorem because I’m actually needed to prove it,” and added that “Human mathematicians are forevermore dethroned as the main theorem-proving entities on planet earth.”
OpenAI and the summer of 2026
The September 8 announcement landed after a summer in 2026 when artificial intelligence models were offering proofs of decades-old mathematical conjectures every week. That pace helps explain why this claim drew such fast reaction. Mathematicians were not hearing about one isolated result. They were watching a field they treat as highly rule-bound get pushed by machines at a speed that unsettled even seasoned researchers.
The immediate fight was over authorship and standards. A solved problem is not the same thing as a proof that the discipline will accept without argument. In mathematics, the missing piece is not just whether an answer exists. It is whether the reasoning is rigorous enough, and whether people judge the machine’s output by the same standards they would demand from a human.
Ken Ono in Berkeley
Two days later, Ken Ono spoke at the University of California, Berkeley, after taking a leave of absence from the University of Virginia to work at Axiom Math. He told about 150 students, postdocs, and professors that, “You might be graduating into a profession that might not even exist, or that will be very different than what you expected,” and added, “You need to brace.”
The room did not react with applause. It reacted with anger and frustration. The talk ran more than two hours after its allotted 50 minutes because of an extended question-and-answer session, and many students lingered afterward to vent and console one another.
Friends and AI, mathematicians
The split in reaction was stark. Friends and family saw the AI breakthroughs as amazing discoveries. Mathematicians in the room sounded closer to grief, fear, and anger. That divide matters because it shows the announcement is being received not as a tidy automation story, but as a challenge to professional identity, training, and status.
The open question is still the one that will decide how far this goes: which specific math problem OpenAI says it solved, and what proof standard was used to back the claim.
