Mathematician Claims Claude AI Solved 167-Year-Old Riemann Hypothesis
A mathematician at Anthropic claims that the Claude AI model helped solve the 167-year-old Riemann hypothesis, producing a 250,000-page proof that awaits global verification.

A mathematician working at Anthropic claims to have solved the Riemann hypothesis—one of the greatest and oldest unsolved problems in mathematics—with the assistance of the Claude AI model. Levent Alpöge announced the development, which could represent a historic breakthrough if independently verified by the global mathematical community.
The Riemann Hypothesis and the Million-Dollar Challenge
Introduced in 1859 by German mathematician Bernhard Riemann, the hypothesis deals with the distribution of prime numbers. For over a century and a half, generations of mathematicians attempted to prove it without success. In the year 2000, it was selected as one of the seven Millennium Prize Problems, earning it the reputation as the holy grail of mathematics alongside a one-million-dollar bounty offered by the Clay Mathematics Institute for its solution.
Alpöge published his proposed proof on his personal website using Lean, a programming language that allows computers to check mathematical arguments step by step for logical validity. Mathematicians worldwide immediately began reviewing the arguments, although the proof has not yet been independently verified. Alpöge described the AI-generated text as astonishingly long and exceptionally complex, spanning more than 250,000 pages and forming a massive, unreadable volume that complicates manual inspection.
Broader Context and Timing of the Announcement
If confirmed, this development would mark a monumental milestone in mathematical problem-solving using artificial intelligence. It could potentially become the second instance of AI assisting in solving a Millennium Prize Problem, following an announcement by OpenAI in September regarding the resolution of a different problem from the list using millions of dollars in computing power. To date, only one of the seven Millennium Prize Problems has been officially solved by a human mathematician.
Alpöge noted that the breakthrough occurred late at night between Sunday and Monday, just days before Anthropic went public. Due to the tight timing surrounding the public offering, which he described as terrible, he contacted Anthropic CEO Dario Amodei to request permission to publish the finding, stating that he simply could not live with himself if he waited any longer.





