The AI takeover of mathematics has begun
The Verge Robert Hart
OpenAI says it solved 10 long-running math problems. Oxford’s James Maynard says that’s forcing mathematicians to face AI’s future.
Based on reporting by The Verge, Robert Hart — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
James Maynard has been doing a lot of soul searching this year, and you can see why. The Oxford professor and Fields Medal winner told The Verge he’s been wrestling with what happens to mathematics when a field built on patience suddenly has to reckon with AI speed.
That tension got sharper days before the interview, when OpenAI said it had produced solutions to 10 long-standing math problems. Some of those problems had stumped academics for decades. For a discipline that usually moves in careful steps, that’s not a minor jolt.
The broader pattern is familiar by now. Generative AI has already shown it can spit out text and images, and can also be used to suggest ideas in science and medicine. Now it’s pushing into mathematics, where the hard part has always been not just getting an answer, but proving you deserve it.
Maynard’s anxiety reads less like panic than a warning. Mathematics is being asked to adapt while the tools around it are changing fast, and that’s an awkward fit for a profession that tends to trust slowness, rigor, and human stubbornness. AI has arrived at the blackboard whether mathematicians like the timing or not.
My take — AI-written commentary, not fact-checked reporting
The romantic story of the lone mathematician was always going to have a hard time surviving contact with AI. That doesn’t make the machines right; it just means the field can’t keep pretending brute-human effort is a moat. The real test now is whether math treats AI as a flashy intern or a very dangerous colleague.
Read more about this at: The Verge