Lean Experts Certify Riemann Hypothesis Proof as Correct - US News Pulse
Independent experts have certified a proof of the Riemann Hypothesis as correct, one day after an Anthropic mathematician announced he had solved the most famous unsolved problem in mathematics with the help of the Claude AI model. The Lean Focused Research Organization and the Mathlib Initiative said in a joint statement on Saturday, November 21, 2026: “We have full confidence that Alpöge’s proof of the Riemann Hypothesis is correct.” The statement added: “We hereby certify that both the statement and the proof of the Riemann Hypothesis are correct.” The certification follows verification of 17 million lines of Lean code.
Key Points
- The Lean Focused Research Organization and the Mathlib Initiative certified Levent Alpöge’s Riemann Hypothesis proof as correct on Saturday, November 21, 2026.
- The community that maintains the Lean proof-checking software finished verifying 17 million lines of Lean code.
- The certification covers both the formal statement and the computer-checked proof.
- The full written argument, put at more than 250,000 pages, is still to come, and mathematicians must still read and review it.
- The Clay Mathematics Institute has not changed its position on its $1 million Millennium Prize.
What Happened
Alpöge announced on Friday, November 20, 2026, that he believes he and Claude discovered a proof of the Riemann Hypothesis, releasing a four-page note and a version of the proof written in Lean, a language that checks mathematical arguments step by step. Within hours, independent mathematicians downloaded the code and ran it on their own machines. Andrew Sutherland of MIT was among those who compiled the Lean proof separately and reported no issues. Scientific American reported that researchers were now almost certain the proof was real, pending a final check that the code proves the hypothesis as mathematicians actually define it.
That check is what Saturday’s certification appears to settle. The statement certifies both parts: the statement — that the formal claim in the code really is the Riemann Hypothesis — and the proof — that every logical step in the code holds. The two bodies behind it build and maintain Lean itself, so their word carries particular weight.
Background: A 167-Year-Old Question About Prime Numbers
Bernhard Riemann proposed his hypothesis in 1859, in a paper of just nine pages. It concerns the zeta function, a mathematical object that encodes deep information about the prime numbers — the indivisible building blocks of all whole numbers. Riemann predicted that all of the zeta function’s non-trivial zeros lie on a single vertical line, the critical line, where the real part equals one half.
If true, it would pin down how the primes are distributed, settling hundreds of results in number theory that currently assume the hypothesis. That is why the Clay Mathematics Institute named it one of its seven Millennium Prize Problems in 2000, each carrying a $1 million award.
The Riemann Hypothesis Proof, Verified Line by Line
What makes this claim different from past announcements is the Lean code. A traditional proof is a written argument that other mathematicians read and check by hand, a process that can take years for a result of this size. A Lean proof is different: the computer verifies every logical step, and if the code compiles, the argument is formally correct — provided the statement it proves is the statement mathematicians mean.
That proviso is exactly what the certification addresses. The certification says the statement matches the classical hypothesis and the proof is correct. What it does not do is replace human understanding: the full written argument, put at more than 250,000 pages, has not yet been produced, and mathematicians still need to read and review it.
Reactions
Alpöge has been candid about the limits of his own grasp of the argument. He said he had not yet mastered all of the mathematics involved and plans to write a full version later. He said the work came together late on a Sunday night, and he consulted Anthropic chief executive Dario Amodei before announcing, a day after Anthropic’s record initial public offering. Anthropic declined to comment on the announcement.
The same day brought a second surprise: a proof of the broader Generalized Riemann Hypothesis by Alessandro Masullo, produced entirely by a human mathematician without AI assistance, was circulated by Ravi Vakil, president of the American Mathematical Society. Alpöge’s claim also covers the broader Generalized Riemann Hypothesis.
Why It Matters
If the proof holds, it would be one of the great mathematical events of the century. The Riemann Hypothesis sits at the centre of number theory, and settling it would confirm or close off a vast web of results built on the assumption that it is true.
The AI angle matters just as much. Sceptics have long argued that AI can pattern-match but cannot do genuinely original mathematics. A computer-checked proof of the most famous open problem in the subject, produced by a mathematician working with Claude, would be a powerful answer — and a sign of how fast AI-assisted research is advancing. The certification makes that answer considerably harder to dismiss.
What Happens Next
The immediate next step is the full written version Alpöge has promised, which will let mathematicians read the argument in the traditional way and subject it to peer review. The Clay Mathematics Institute requires a proof to be published in a refereed journal and to survive two years of scrutiny before the $1 million prize can be awarded, and it has not changed its position.
The mathematics community will now spend months testing every line of the argument. A computer check proves the logic; only human review can show whether the ideas behind it are sound and understood. Until that review is done, the certification is a powerful endorsement — not a closing verdict.
Frequently Asked Questions
What is the Riemann Hypothesis?
A conjecture from 1859 predicting that all non-trivial zeros of the zeta function lie on the critical line. It governs how prime numbers are distributed, and it is the most important open question in pure mathematics.
Who certified the proof?
The Lean Focused Research Organization and the Mathlib Initiative, the bodies behind the Lean proof-checking software, issued a joint statement on November 21, 2026, certifying that both the statement and the proof are correct.
What is Lean, and why does it matter?
Lean is a programming language that checks mathematical proofs step by step. If code compiles without errors, the argument is formally correct — the remaining question was whether it proves the hypothesis as mathematicians define it.
Is the proof now fully accepted?
Not yet. The certification covers the formal statement and the computer-checked proof. Mathematicians still need to read and review the full written argument, put at more than 250,000 pages, and peer review has not begun.
What about the $1 million prize?
The Clay Mathematics Institute has not changed its position. Its Millennium Prize requires publication in a refereed journal and two years of scrutiny after that, so no prize decision is imminent.
Read more: Claude AI Helped Prove the Riemann Hypothesis, Mathematician Claims; Anthropic Closes First Trading Day at $97.17, Up 17% From $83 IPO Price; Strict school phone bans lift maths and reading scores, Harvard study finds.
Leave a Comment