Tyrian Trade

Anthropic Researcher Claims Claude Helped Prove the 167-Year-Old Riemann Hypothesis

Anthropic mathematician Levent Alpöge has published a claim that he and the company's Claude AI model developed a proof of the Riemann Hypothesis, one of mathematics' most important unresolved problems. The proposed solution has not yet been independently verified.
The Riemann Hypothesis, formulated by Bernhard Riemann in 1859, concerns the location of the nontrivial zeros of the Riemann zeta function. It is closely connected to the mathematical understanding of prime-number distribution and is one of the seven Millennium Prize Problems established by the Clay Mathematics Institute.
Alpöge released a four-page explanatory note on November 20, saying he believed he had discovered a proof while working with Claude. His account describes a research effort beginning in early October and progressing through multiple mathematical ideas over approximately six weeks.
The researcher also reported that parts of the argument had been formalized using Lean, a proof assistant capable of checking mathematical statements according to precise logical rules. However, formal verification of a computer-encoded theorem would not settle the wider claim unless independent experts confirm that the formal statement corresponds to the original Riemann Hypothesis and that the relevant assumptions are valid.
The announcement follows earlier AI-assisted work on related problems in analytic number theory. Those partial advances attracted attention but did not establish the full conjecture. Alpöge's latest claim is substantially stronger and therefore requires particularly careful scrutiny.
Only one of the seven Millennium Prize Problems has been officially resolved: the Poincaré Conjecture, proved by Grigori Perelman. The Clay Mathematics Institute offers a $1 million prize for an accepted solution to each remaining problem, subject to its verification and recognition requirements.
The potential importance of a verified proof would extend well beyond Anthropic. It could represent a historic development in mathematics and demonstrate a significant advance in AI-assisted reasoning. At present, however, the claim remains a proposed result rather than an established solution.
Mathematicians will need to examine the argument, its formalization and any supporting documentation before reaching a conclusion. The existence of a published proof claim is newsworthy, but it should not be confused with independent confirmation that the 167-year-old conjecture has been solved.

Anthropic Researcher Claims Claude Helped Prove the 167-Year-Old Riemann Hypothesis

Loading…