Skip to content
The Daily Circuit
AI News

Anthropic Mathematician and AI Claim Solution to 167-Year-Old Riemann Hypothesis

Anthropic mathematician Levent Alpöge has released a Lean-compilable proof of the Riemann hypothesis developed with Claude, alongside a separate human-only proof by Stanford postdoc Alessandro Masullo.

By The Daily Circuit AI Desk (automated, AI-generated)

2 min read

Abstract digital representation of mathematical equations and AI neural networks.

AI-generated image (flux-1-schnell), not a photograph of the events described.

Key facts

  • Anthropic mathematician Levent Alpöge claimed a proof of the Riemann hypothesis developed with AI model Claude [S1][S2].[1][2]
  • The Riemann hypothesis dates back to 1859 and is one of the Clay Mathematics Institute's Millennium Prize problems [S1][S2].[1][2]
  • Alpöge's solution is expressed in Lean and has been checked and compiled successfully by independent sources [S1].[1]
  • Stanford postdoc Alessandro Masullo released a human-only proof of the General Riemann hypothesis following 12 years of work [S1].[1]
  • Clay Mathematics Institute rules require a proposed solution to withstand at least two years of scrutiny before official prize consideration [S2].[2]

Levent Alpöge, a mathematician at Anthropic, has claimed a breakthrough solution to the 167-year-old Riemann hypothesis developed with the help of AI model Claude [1][2]. The announcement coincides with Anthropic's public market debut as a frontier AI lab [1].

The Riemann hypothesis, dating back to 1859, is one of the seven Millennium Prize problems administered by the Clay Mathematics Institute, which offers a $1 million bounty for each verified solution [1][2]. The hypothesis conjectures a hidden structure underlying the distribution of prime numbers [1]. Alpöge stated in an essay and on his website that working with Claude enabled him to discover a proof that also extends to the Generalized Riemann Hypothesis [1][2]. He described messaging Anthropic CEO Dario Amodei late Sunday night to discuss how to announce the findings [1][2].

Verification and Community Response

Alpöge's solution is expressed in Lean, an open-source coding language used to express mathematical ideas as code to allow formal verification [1]. In a Zulip chat, experts including MIT mathematician Andrew Sutherland and number theorist Chris Birkbeck examined the arguments, with independent sources successfully confirming that Alpöge's Lean proof can be compiled successfully [1]. However, mathematicians note that it will take time to fully understand the proof and explain it human-intelligibly, and the solution has not yet been publicly accepted by the Clay Mathematics Institute [1][2]. Under the institute's rules, a proposed solution must undergo at least two years of scrutiny and general acceptance in the global mathematics community [2]. Anthropic had previously noted in August that it was working on the Riemann hypothesis with Claude, though it stated at the time that it did not expect the techniques to lead to a proof [2]. Anthropic has not formally commented or involved itself in Alpöge's announcement [2].

Competing Human Proof and Navier-Stokes Context

In a parallel development, Stanford professor and American Mathematics Association president Ravi Vakil posted a proof of the General Riemann hypothesis written by Stanford postdoc Alessandro Masullo [1]. Vakil stated that Masullo's proof is entirely human with no AI involvement, representing the culmination of 12 years of work [1]. Without being expressed in Lean, experts note it is too early to determine whether Masullo's proof checks out [1]. Additionally, Alpöge and New York University mathematician Tristan Buckmaster released another proof related to the Navier-Stokes equations [1][2], following earlier industry developments in mathematical problem-solving by frontier AI labs [2].

Context

The Riemann hypothesis is one of seven Millennium Prize problems established by the Clay Mathematics Institute in the year 2000, carrying a $1 million bounty [2]. Proposed originally in 1859 by Bernhard Riemann, it deals with prime numbers and the zeroes of the Riemann zeta function [1][2]. Formal verification languages like Lean allow complex mathematical ideas to be coded and checked systematically [1].

Why it matters

The intersection of artificial intelligence and advanced mathematics is shifting traditional academic paradigms, raising questions about verification, human-intelligible explanations, and the pace of discovery in fundamental science.

What could happen next

AI-generated analysis based on the sources below. Not a forecast or recommendation.

The integration of AI tools like Claude into advanced mathematical research may accelerate the exploration of complex theoretical problems, though it places a heavy verification burden on the human mathematical community.

Risks and open questions

It remains unconfirmed whether Alpöge's proof is fully understood by human mathematicians or if it will withstand the required two-year scrutiny period by the Clay Mathematics Institute. Masullo's human-authored proof also requires extensive review.

Sources

  1. 1.TechCrunch (primary source): AI claimed a solution to the 167-year-old Riemann hypothesis — and a human did, too, 21 November 2026
  2. 2.The Verge: Anthropic mathematician claims to have solved prestigious Millennium Prize problem with Claude, 20 November 2026

How this article was produced. This story was researched and written by an automated system using Google’s Gemini models, working only from the sources listed above. Automated checks verified that every citation, quote and figure matches those sources before publication. It was not written by a human reporter. Editorial policy · Report an error

  • Artificial Intelligence
  • Anthropic
  • Mathematics
  • Claude