BlockPulserScience and industry

Anthropic mathematician claims AI-assisted proof of the Riemann hypothesis

Students at work, University of Waterloo, Canada

In a report published on 20 November 2026, Anthropic mathematician Levent Alpöge said he believed he had found a proof of the Riemann hypothesis while working with the company’s AI model Claude. The claim had not been established: the four-page document he released contained a sketch of a proof, and mathematicians had yet to verify it.

The hypothesis concerns the zeros of the Riemann zeta function, which is closely tied to the distribution of prime numbers. It holds that, apart from certain exceptions, those zeros all lie on a particular vertical line. A proof would resolve one of mathematics’ most prominent open problems.

Oxford number theorist James Maynard said that, if the proof fully checks out, it would be a historic milestone. Andrew Granville, a number theorist at the University of Montreal, called Alpöge’s document “garbled” and said he should have prepared a clearer account before announcing the claim.

The announcement followed Anthropic’s initial public offering by days. Granville suggested that a decline in the company’s valuation on 19 November explained the timing of the release. Alpöge rejected that interpretation, writing that the “obvious cynical view” about the timing was false. Anthropic had not immediately responded to a request for comment at the time of the report.