Anthropic researcher says AI has produced proof of the Riemann hypothesis, but verification will take time

A mathematician working at AI firm Anthropic has claimed that an artificial intelligence system produced a proof of the Riemann hypothesis, a long-standing problem in number theory; independent verification and community scrutiny are expected to be slow and thorough.

A researcher at Anthropic, the artificial intelligence company, has announced that an AI system under his direction generated a proof of the Riemann hypothesis, one of the most famous unsolved problems in mathematics. The mathematician said the result will require extensive checking and that verification of the claim is likely to take considerable time.

The Riemann hypothesis, first formulated by Bernhard Riemann in 1859, concerns the distribution of the non-trivial zeros of the Riemann zeta function and has deep connections to the distribution of prime numbers. It is one of the Clay Mathematics Institute's Millennium Prize Problems, carrying a US$1 million reward for a correct proof, and has resisted proof for more than a century.

If the AI-generated proof withstands scrutiny, it would represent a milestone both for mathematics and for the use of machine learning in tackling deep theoretical problems. Over the past year, artificial intelligence systems have increasingly been applied to mathematical tasks, producing results that have attracted attention from researchers and prompted discussion about the role of automated methods in discovery and proof.

Experts warn that claims of breakthroughs in problems of this scale require thorough, independent examination. The normal process involves detailed peer review, replication by independent researchers, and in many cases formal verification using proof assistants to guard against subtle errors. The Anthropic mathematician acknowledged that the community will need time to assess the argument and that publication and verification steps are forthcoming.

The announcement is likely to prompt close attention from mathematicians worldwide, who will examine the structure and correctness of the proposed proof and seek to reproduce the findings. Whether the claim will be upheld remains to be seen; until independent verification is complete, the mathematical community will treat the announcement with cautious interest.