Fields Medalist Maynard on "AI Cracking the Riemann Hypothesis": I Believe the Proof Is Correct, but a Result Can Only Be Recognized Once It Is Understood
IT之家 November 22 news: On November 20 local time, Anthropic mathematician Levent Alpöge announced that he had collaborated with the AI model Claude to crack one of the most famous problems in mathematics — the Riemann Hypothesis.
According to previous IT之家 reports, on November 21, the Lean and Mathlib communities issued a joint statement that a proof of the Riemann Hypothesis given in the form of a Lean library had passed verification, and the code for the proof is about 17 million lines.
According to 每日经济新闻 reports, 2022 Fields Medalist and University of Oxford mathematics professor James Maynard said in an interview that he believes the proof is correct, but stressed that "understanding why a theorem holds is far more important than merely proving that it holds." Only when someone truly understands this result and can explain it clearly to others will the proof have a chance of gaining full recognition from the mathematics community.
Maynard is an important mathematician in the field of contemporary analytic number theory, and has long devoted himself to research on mathematical problems such as the distribution of primes. In 2022, he won the Fields Medal for related research results.

Because this result has already been formally verified, and this verification has in turn been independently checked. In addition, Alpöge has a reliable track record in making such major claims in the past. Taken together, Maynard said he is very confident that the result is correct.
Maynard said that although the Riemann Hypothesis does have some direct mathematical corollaries, the reason this problem is considered so important is mainly that people have long expected that explaining this profound internal structure in the distribution of primes would require developing new mathematical theories.
Before we truly understand the ideas and overall framework in the proof, this result will likely fail to bring about the transformation people expect. However, I believe that over time we will eventually truly understand this proof, and that will certainly completely change our understanding of the relevant mathematical problems.
Before we truly understand the ideas and overall framework in the proof, this result will likely fail to bring about the transformation people expect.
However, I believe that over time we will eventually truly understand this proof, and that will certainly completely change our understanding of the relevant mathematical problems.
- "AI Cracks the Riemann Hypothesis? Anthropic Mathematician Claims to Have Completed the Proof with Claude"
- "Anthropic Mathematician's AI Cracks the Riemann Hypothesis? Lean Community Passes Verification, Dozens Report Codebase Compiles Correctly"
Related Articles
- Anthropic Mathematician's AI Cracks the Riemann Hypothesis? Lean Community Passes Verification, Dozens Report Codebase Compiles Correctly
- Was the Riemann Hypothesis Really Proved by Claude? Anthropic Mathematician Only Released Lean Code, Written Proof Exceeds 250,000 Pages
- AI Cracks the Riemann Hypothesis? Anthropic Mathematician Claims to Have Completed the Proof with Claude
- Fields Medalist Deng Yu Warns: Humanity Cannot Give Up Independent Thinking and Become Dogs for AI
- Fields Medalist Deng Yu Comments on AI Companies' Anti-Distillation: You Cannot Support Distillation Only When You Distill All Human Knowledge
- 25 Fields Medalists Jointly Warn That "AI Is Seriously Misaligned in Mathematics," Deng Yu Jokes About Retiring to Go Home and Write Yuri Novels