Alpöge, who works at Anthropic, said in an essay that he had asked CEO Dario Amodei last Sunday how to announce his result. The hypothesis, first posed in 1859, concerns a proposed pattern behind the distribution of prime numbers and is one of the Millennium Prize problems.
The proof is written in Lean, an open-source language used to check mathematical arguments. Mathematicians discussing it on Zulip have begun examining its technical details. Chris Birbeck, a number theorist at the University of East Anglia, described an argument involving automorphy lifting and several results in number theory. The proof has not yet been widely understood, although the source report says many observers are confident the formalized result will hold.
Alpöge acknowledged that he had not provided an explanation that was readily intelligible to other mathematicians. He wrote that his longstanding fascination with the problem made him feel he “could not live” with waiting to release the finding. In describing work with Anthropic’s Claude, he called the process a “stream-of-consciousness jaunt” and said the model “flew, and danced, and sparkled near that finish line.” He has committed to helping people understand the proof.
Alpöge and NYU mathematician Tristan Buckmaster have also released another proof concerning the Navier-Stokes equations, problems that helped spur a recent race between frontier AI labs and mathematicians to solve major open questions.
Separately, Ravi Vakil, a Stanford professor and president of the American Mathematics Association, posted Masullo’s proof of the General Riemann Hypothesis. Vakil said it was written without AI and that the plan to bring Masullo to Stanford for a year to polish it had given way to releasing the current draft. Masullo said the work represented 12 years of effort and urged mathematicians discouraged by the current climate to remember: “we are still here. We are very much alive. We are not going anywhere.”

Comments (0)
No comments yet. Be the first!