Dallas – A mathematician affiliated with Anthropic, the artificial‑intelligence firm that just completed its initial public offering on Nasdaq, announced that the company’s Claude model helped him develop a proposed proof of the Riemann Hypothesis, a problem that has challenged mathematicians since 1859.
Claim and Methodology
Levent Alpöge, who posted a four‑page statement on his personal website on Friday, said he believes he has discovered a proof of the hypothesis while working with Claude. He also asserted that the same approach yields a proof of the Generalized Riemann Hypothesis for a class of Dirichlet L‑functions.
Alpöge released the argument in Lean, a formal proof language that allows software to verify each logical step against defined axioms and definitions. An experimental run of the proof generated more than 250,000 pages of verification output, according to a footnote in his statement.
Independent Checks
Andrew Sutherland, a mathematician at the Massachusetts Institute of Technology, used Claude Code to run a series of checks on the Lean file. In a brief report posted Friday afternoon, Sutherland noted that the source compiled without errors, that five main theorems rely only on Lean’s standard axioms, and that a separate proof‑checking program accepted the full exported proof. Two additional checks were still in progress at the time of his update.
Context and Significance
The Riemann Hypothesis predicts a precise location for the non‑trivial zeros of the Riemann zeta function, which in turn tightly bounds the error term in the average distribution of prime numbers. While many aspects of prime distribution are understood, the hypothesis remains the central unsolved question in analytic number theory.
The Clay Mathematics Institute has offered a $1 million prize for a solution that meets its criteria: publication in a recognized venue, at least two years of acceptance by the global mathematics community, and broad peer validation. The institute still lists the hypothesis as unsolved.
Anthropic’s Recent IPO
Anthropic announced in October that it intended to go public, and the company confirmed on Friday that its shares began trading on Nasdaq on Thursday. The IPO marks a significant milestone for the AI firm, which has positioned Claude as a competitor to other large‑language models in the market.
While the mathematical community will likely scrutinize Alpöge’s claim over the coming months, the involvement of a leading AI system in such a high‑profile problem underscores the growing role of artificial intelligence in advanced research.
Next Steps
Alpöge indicated that he plans to prepare a conventional, peer‑reviewable manuscript of the proof. Further verification by independent mathematicians and formal proof assistants will be essential before the claim can be accepted as a definitive solution.
Original reporting: The Dallas Express — read the source article.