Mathematician Levent Alpöge and AI Claim Riemann Hypothesis Proof

Anthropic mathematician Levent Alpöge claimed on November 20 that he and the artificial intelligence model Claude have successfully proved the Riemann hypothesis, a legendary open problem in mathematics.

The announcement arrived just a day after Anthropic completed its initial public offering on the stock market. The company debuted with a valuation of US$1.5 trillion, though shares declined on November 19, according to Nature reporting.

First formulated by German mathematician Bernhard Riemann in 1859, the hypothesis deals with the distribution of prime numbers through the Riemann zeta function. The Clay Mathematics Institute designated it as one of seven Millennium Prize Problems in 2000, offering a $1 million bounty for a solution, as Scientific American noted.

Mathpocalypse: Riemann Hypothesis "solved" by Claude AI and Anthropic mathematician

“I believe I have, working with Claude, discovered a proof of the Riemann Hypothesis,” Alpöge wrote in a statement posted to his website, as reported by Nature.

Evaluating the AI-Assisted Proof

Alpöge released his findings in computer-checkable Lean code rather than a traditional written manuscript. He explained that an attempt to produce a conventional prose explanation grew beyond 250,000 pages, prompting him to share the argument through rigorous verification software instead, as reported by the New York Post.

“Claude far outpaced me near the finish line,” Alpöge wrote, adding that the artificial intelligence “flew, and danced, and sparkled near that finish line.”

Reactions from the mathematical community were mixed regarding both the presentation and the timing of the release. Andrew Granville, a number theorist at the University of Montreal, criticized the documentation provided by his colleague.

“What he released is garbled,” Granville told Nature, suggesting that the public rollout was tied to the company’s recent stock market debut. Alpöge rejected that assertion in his published statement, writing, “There is an obvious cynical view, to do with the timing, but it is false.”

Independent Expert Verification

Other mathematicians voiced cautious optimism while emphasizing the scale of the milestone. James Maynard, a Fields Medal winner at the University of Oxford, said that the claim represents a major development if verified.

“If this fully checks out, it is clearly a historic milestone in mathematics,” Maynard told Nature.

Massachusetts Institute of Technology mathematician Andrew Sutherland said that he had independently compiled the result in Lean. Sutherland described the event as a historic moment for the field, according to Scientific American.

Math equations with purple digital concept
Photo: Scientific American

Simultaneously, Stanford mathematician and American Mathematical Society President Ravi Vakil announced on his blog that another mathematician, Alessandro Masullo, had submitted a separate, entirely human-generated proof of the Generalized Riemann Hypothesis that had been under review for months, as reported by OfficeChai.

Questions About Prime Number Mathematics

FAQ:

What is the Riemann hypothesis?
Proposed by Bernhard Riemann in 1859, it is an unproven mathematical conjecture concerning the zeroes of the Riemann zeta function and the distribution of prime numbers among whole numbers.

How did Levent Alpöge present his proof?
Rather than publishing a traditional prose manuscript, Alpöge released the argument in Lean code, a computer program designed to verify mathematical proofs for logical errors.

What did Anthropic say about the announcement?
Anthropic did not respond immediately to requests for comment from Nature following the release of the proposed proof.

Anthropic was founded in January 2021 by former OpenAI employees, including siblings Daniela Amodei and Dario Amodei, following disagreements over AI safety. The company developed the large language model Claude, which Alpöge credited with outpacing him near the finish line of the proof.

Anthropic completed its initial public offering in November 2026, the same month as the Riemann hypothesis announcement. In May 2026, a Series H capital raise valued the public benefit corporation at $965 billion, making it the world’s most valuable AI pure play.

Anthropic did not respond immediately to Nature’s request for comment.