AI claimed a solution to the 167-year-old Riemann Hypothesis — and a human did, too

Anthropic becomes a public company while Levent Alpöge solves the Riemann Hypothesis Levent Alpöge, a mathematician at Anthropic, has announced a solution…

By Vane November 20, 2026 2 min read
AI claimed a solution to the 167-year-old Riemann Hypothesis — and a human did, too


Anthropic becomes a public company while Levent Alpöge solves the Riemann Hypothesis

Levent Alpöge, a mathematician at Anthropic, has announced a solution to the Riemann Hypothesis. This result dates to 1859 and is one of the Millennium Prize problems. The same week, his employer became the first publicly traded frontier AI laboratory.

Alpöge described the timing as unfortunate in an essay outlining his findings. He wrote that he messaged Anthropic CEO Dario Amodei last Sunday night to ask how to announce a Lean-compilable solution.

The Riemann Hypothesis conjectures that a hidden structure governs the appearance of prime numbers. If the result is correct, it confirms a fundamental pattern in number theory.

Mathematicians are currently reviewing the work on a Zulip chat. Chris Birkbeck, a number theorist at the University of East Anglia, noted that the cohomology bound relies on an automorphy-lifting argument.

The cohomology bound is proved by what looks like an automorphy-lifting argument. It genuinely uses the existence of Galois representations attached to regular algebraic, polarized cuspidal automorphic representations of GL_n over CM fields, along with base change, Jacquet–Shalika, Clozel’s purity, Bernstein–Zelevinsky, the main theorem of CM, Deligne–Ribet and more.

Birkbeck added that he left mathematics at college calculus and cannot decode the proof.

Twenty readers reacted with shock to that comment.

The community will take time to verify the proof, but most appear confident the formalized Lean result is correct. Lean is an open source language used to verify mathematical proofs.

Alpöge expressed guilt about releasing the proof without a fully human-intelligible explanation. However, his obsession with the problem since childhood made him feel he could not wait to release the finding.

He described a stream-of-consciousness process of finding the argument pieces through confusion with Claude. He wrote that the model flew, danced, and sparkled near the finish line.

Humans must still understand and explain the result to each other. Alpöge has committed to pursuing this work.

Alpöge and collaborator Tristan Buckmaster from NYU also released a proof related to the Navier-Stokes equations. These problems helped spark a race between frontier labs and mathematicians to solve outstanding issues this summer.

Ravi Vakil, a Stanford professor and president of the American Mathematics Association, posted a proof of the General Riemann Hypothesis by Stanford postdoc Alessandro Masullo. Vakil noted the work is entirely human, with no AI involvement. The hope was to bring Masullo to Stanford for a year to polish the writing and announce it after review. Vakil said the current draft is best published now.

In a brief introduction to his proof, which Masullo said is the result of 12 years of work, he wrote that if any message is meant to send to mathematicians discouraged by the current climate, it is this: we are still here. We are very much alive. We are not going anywhere.

What it means

These developments show that AI tools can assist in verifying complex proofs or generating candidate arguments, but human mathematicians remain essential for understanding, validating, and communicating the results.


Scroll to Top