Has the Riemann Hypothesis Finally Been Solved? A Mathematician Says He and an AI Have the Proof
The most famous unsolved problem in mathematics may — or may not — have just been solved. On 20 November 2026, Levent Alpöge, a mathematician at the artificial-intelligence company Anthropic, announced that he believes he and the company's AI system, Claude, have discovered a proof of the Riemann hypothesis. The claim is reported by Nature, but independent mathematicians have not yet verified it — and some leading experts are urging caution.
The Riemann hypothesis is about prime numbers: 2, 3, 5, 7, 11 and the rest, the building blocks of all whole numbers. In 1859, the German mathematician Bernhard Riemann — born 200 years ago this September — guessed a precise pattern in how primes are scattered among the numbers. His guess involves a mathematical object called the Riemann zeta function, and he predicted that all of its important "zeroes" lie on one particular vertical line. If true, it would mean primes are spread out about as evenly as they possibly could be. Nobody has proved it in 167 years, and the Clay Mathematics Institute offers a $1 million prize for a solution.
Alpöge says the new proof also covers the generalised Riemann hypothesis, a wider version of the same idea. Rather than publishing a conventional paper, he released the argument in Lean, a computer language that checks mathematical reasoning step by step, together with a four-page sketch of the ideas. He acknowledged that the argument is extraordinarily long and difficult to read.
The reaction from mathematicians has been a mix of excitement and scepticism. James Maynard, a Fields Medal winner at the University of Oxford, told Nature: "If this fully checks out, it is clearly a historic milestone in mathematics." But Andrew Granville, a number theorist at the University of Montreal, said the four-page sketch was "garbled" and that Alpöge should have prepared a clearer write-up before announcing — especially since the announcement came just days after Anthropic's stock market flotation, whose valuation had dropped on 19 November. Alpöge has denied that the timing was connected to the flotation.
This is the second big AI-and-primes story in as many months: in October, OpenAI announced that its AI had proved a weaker version of the same problem, the so-called quasi-Riemann hypothesis, which we covered here at the time. The new claim goes much further — if correct, it would settle the whole question, not just a weaker cousin. But as one expert put it, checking the computer-checked proof is only the beginning; mathematicians would need to understand how it works before broadly accepting it, a process that could take months or even years.
For families, this is a wonderful moment to talk about how maths actually works. A proof is not a claim or a headline — it is an argument that other mathematicians can check, step by step, until they are convinced. Even if this claim turns out to have a flaw, the story shows how AI is changing the way mathematicians work, and why prime numbers still matter: they guard our bank accounts, our messages and our secrets every day. If your child is fascinated by numbers, patterns and puzzles, you can find a maths tutor to explore them with, post a tutoring request to compare tutors, or browse GCSE maths resources on TES.
If your child loves prime numbers and big puzzles, a first lesson in maths can start with the Riemann hypothesis and end with the patterns hiding in the numbers themselves: Book your first lesson
Sources:
- "Mathematician says he and AI solved Riemann hypothesis — the field's biggest outstanding challenge", Nature, 20 November 2026: https://www.nature.com/articles/d41586-026-03812-7?error=cookies_not_supported&code=1261f801-d2c9-4d31-a71d-a250e3c35f2c
- "Anthropic mathematician claims AI helped solve a 167-year-old math mystery", IBM Think, 20 November 2026: https://www.ibm.com/think/news/anthropic-claude-riemann-hypothesis-math-proof
Curious about maths? Post a request on London Tutors and compare verified maths tutors — online, in person or in a group class.