International Edition
Latest News
Business

Anthropic Mathematician Claims AI Helped Solve a 167-Year-old Math Mystery | IBM

Anthropic mathematician Levent Alpöge said on Friday he had used the company’s Claude model to produce a proof of the Riemann hypothesis, a 167-year-old problem that carries a $1 million Millennium Prize. The claim, posted on Alpöge’s website…

Anthropic Mathematician Claims AI Helped Solve a 167-Year-old Math Mystery

Anthropic mathematician Levent Alpöge said on Friday he had used the company’s Claude model to produce a proof of the Riemann hypothesis, a 167-year-old problem that carries a $1 million Millennium Prize.

The claim, posted on Alpöge’s website with the argument formalized in the Lean proof‑assistant language, would — if verified — settle the Clay Mathematics Institute’s most famous open question and simultaneously prove the far broader Generalized Riemann Hypothesis. Scientific American reported that Alpöge wrote he believes he has “discovered a proof of the Riemann Hypothesis” working with Claude.

University of Lorraine professor Youness Lamzouri, who previously produced a shorter, human‑written proof that 67.25 % of the relevant zeros lie on the critical line, told IBM Think the new claim left him “in awe” and would constitute a “huge achievement” if correct. He cautioned, however, that checking the computer‑generated code is only the first step; mathematicians must understand the underlying ideas before the result can be broadly accepted, a process he said could take months or years. “Mathematics is mainly about understanding why something is true,” Lamzouri told IBM. “We need to understand the ideas in the proof well enough to explain them, teach them and build on them to extend human knowledge.”

The proof spans more than 17 million lines of Lean code, according to the Daily Telegraph, which also said that Alpöge informed Anthropic chief executive Dario Amodei of the result on Sunday — days before the company’s initial public offering. Oxford mathematical physicist Jon Keating called the development “the greatest shock in my lifetime,” while London Institute for Mathematical Sciences researcher Yang‑Hui He said he had thought a proof “inconceivable” in his lifetime.

“All mathematicians see this as the Holy Grail of mathematics,” He told the Telegraph.

Carlo Pagano, a professor at Concordia University and visiting researcher at Google DeepMind, said a first high‑level scan of the argument summary suggests the new ideas could open “brand new avenues of thinking in number theory” regardless of the final verdict. Hector Pasten of the Pontifical Catholic University of Chile compared the potential impact to a “Rosetta Stone” linking previously separate branches of mathematics, Scientific American reported.

Alpöge’s statement acknowledged the emotional weight of the moment. “It is natural to feel that this is an end, in my case a loss of a principal guiding light,” he wrote, “but what I saw during the course of progress so amazed me that I believe it is one dawn across science of something far greater.”

The announcement follows a turbulent year for AI‑assisted mathematics. In September, OpenAI claimed progress on the Navier‑Stokes equations, prompting New York University Tristan Buckmaster to allege the lab had pressured him while he neared his own breakthrough, the Telegraph reported. Alpöge himself was central to that controversy, according to the same report.

Verification of the Lean code is now under way by independent experts. No formal submission to the Clay Mathematics Institute has been announced, and the institute’s prize conditions require publication in a refereed journal and general acceptance by the mathematical community — a process Lamzouri emphasized could extend well beyond the initial code audit.

About the author: Business Desk

Business Desk at Archynewsy.