Anthropic Mathematician Claims Riemann Hypothesis Proof Co-Authored with Claude
Anthropic mathematician Levent Alpoge claims to have proven the Riemann Hypothesis in collaboration with Claude, extending the result to the broader Generalized Riemann Hypothesis and encoding the proof in Lean for formal verification. If confirmed, it would mark the first AI-assisted proof of a top-tier open mathematical problem and a pivotal leap in AI mathematical reasoning.
Related Entities
Full Report
IT之家 November 20 news, according to Scientific American, on the 20th local time, Anthropic mathematician Levent Alpoge claimed that he, in collaboration with the AI model Claude, had cracked one of the most famous problems in mathematics — the Riemann Hypothesis.
Alpoge stated on his personal website: "I believe that I, in collaboration with Claude, have found a proof of the Riemann Hypothesis."
IT之家 note: The Riemann Hypothesis was proposed by German mathematician Bernhard Riemann in 1859, positing that the primes, which appear to be randomly distributed along the number line, actually follow a complex and meaningful pattern.
Since then, the Riemann Hypothesis has become an important guiding principle for the entire field of number theory research, and is also the most famous unsolved problem in mathematics. Due to the long-term lack of clear proof approaches, research progress has been extremely limited, and even top mathematicians rarely spend significant time directly attempting to prove it. It was not until October of this year that among a batch of mathematical research results collectively published by OpenAI, one proof made significant progress in revealing the patterns of prime distribution, giving the first glimpse of a breakthrough for tackling the Riemann Hypothesis.
The long road of verification seems to be nearing its end. Alpoge published the proof completed in collaboration with Claude and encoded it as Lean code.
It is reported that Lean is a programming language capable of step-by-step verification of whether mathematical reasoning is correct. Once confirmed, 2026 will become a year of leapfrog progress in AI mathematical capability.
The report says that AI has gone from previously performing inconsistently, even struggling to reach the level of college students, to now demonstrating capabilities surpassing those of humans, and every branch of mathematics will be affected.
Alpoge claims that his proof not only solves the Riemann Hypothesis but also proves the "Generalized Riemann Hypothesis," which covers a scope far broader than the original conjecture.
The Generalized Riemann Hypothesis implies that the entire mathematical world harbors countless undiscovered deep patterns. If this result is confirmed, the impact on the mathematics community will be immeasurable, and a large number of mathematical conclusions that presuppose the truth of the Generalized Riemann Hypothesis will also be confirmed accordingly.
Alpoge's statement conveys both a sense of sorrow at the end of an era and anticipation for the future. "It is natural to feel that all of this means some kind of ending. For me, it means losing an important guiding light. But the progress I witnessed during the research process has struck me so deeply that I believe the entire scientific community is ushering in the dawn of something far greater."
Alpoge concluded by writing: "I hope this proof allows people to glimpse the wonders of the future, and lets us feel what kind of joy and amazement even the simplest things will bring us."
References
Advertising statement: External redirect links contained in this article (including but not limited to hyperlinks, QR codes, passcodes, and other forms) are used to convey more information and save selection time. Results are for reference only. All articles on IT之家 containing external links include this statement.