Sources checked
Anthropic mathematician claims Millennium Prize solution
Anthropic mathematician Levent Alpoge claims to have solved the Riemann hypothesis, one of mathematics's most notoriously difficult problems, with the help of Claude, cracking a puzzle that has eluded some of the world's best mathematicians for 167 years.
If verified, it would mark another seismic achievement for AI in mathematics and add to the upheaval already unleashed by previous AI-assisted breakthroughs in the field.
Alpoge published the proof on his personal website, writing: "I believe I have, working with Claude, discovered a proof of the Riemann Hypothesis.
In fact, the method easily extends much further, resolving the so-called Generalized Riemann Hypothesis, and I expect, with more time, the Extended Riemann Hypothesis too."
The proof has been formalized in the Lean proof checker, an interactive theorem prover that can verify mathematical arguments mechanically, meaning the logical structure can be checked by software even as human mathematicians begin their review.
Alpoge texted Anthropic CEO Dario Amodei about the solution late Sunday and held off publication until after the company's IPO on Thursday, according to his website.
Anthropic had acknowledged working on the Riemann hypothesis with Claude in August, but said at the time that it did not expect the techniques to yield a proof.
The Riemann hypothesis is one of seven Millennium Prize problems designated by the Clay Mathematics Institute, which offers a one million dollar bounty for each solved problem.
First proposed by Bernhard Riemann in 1859, it concerns the distribution of prime numbers and how the zeroes of the Riemann zeta function are arranged.
The institute notes that the assertion has been checked for the first 10 trillion solutions but a general proof has remained out of reach for over a century and a half.
The proof and how AI contributed
Alpoge says Claude was integral to developing and formalizing the argument.
"It is natural to feel that this is an end, in my case a loss of a principal guiding light, 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," he wrote.
He also believes the method could extend to the Generalized Riemann Hypothesis and the Extended Riemann Hypothesis, describing the argument as containing "an astounding amount of mathematics, far more than I had imagined would be necessary as a student."
This is not the first time Alpoge has claimed an AI-assisted mathematical breakthrough. In July, he said he solved the Jacobean conjecture using Claude.
The rapid accumulation of results from AI-augmented mathematicians has left the field scrambling to digest new material and reassess what it means to do original research.
Verification and the hurdle ahead
The solution has only been published on Alpoge's personal website and has not yet been accepted by the Clay Mathematics Institute.
Under the institute's rules, a proposed solution must be published and withstand at least two years of scrutiny before it can be considered for the prize. During that time, it must receive general acceptance in the global mathematics community.
Anthropic has not commented on the claim. The company did not immediately respond to requests for comment.
In the quarter-century since the Millennium Prize was established, only one of the seven original problems has been officially solved: the Poincare conjecture, a topological problem concerning three-dimensional spheres resolved by Grigori Perelman in 2003.
In September, OpenAI claimed to have cracked another, the Navier-Stokes problem concerning fluid dynamics, though that result has also not received formal recognition.
OpenAI said it spent millions of dollars deploying 10,000 agents to produce its Navier-Stokes solution, though the achievement has been marred by allegations that it used unpublished work from mathematicians who had shared their research with its AI tools.
OpenAI has denied wrongdoing, but the controversy has highlighted the fraught question of credit and priority in AI-assisted mathematics.
Implications for the field
The Riemann hypothesis is regarded as the single most important unsolved problem in number theory. It has been described as the "North Star for the entire field of number theory, and the most famous open problem in math."
Experts previously considered the problem so insurmountable that many did not bother working on it. A verified proof would have far-reaching implications for the understanding of prime numbers, which underpin much of modern cryptography.
The mathematical community will now scrutinize Alpoge's proof. The Lean formalization means that the logical structure can be checked mechanically, eliminating one class of errors.
But the question of whether the argument genuinely proves the Riemann hypothesis, whether the reasoning is novel, and whether the proof holds under standard mathematical assumptions will require human judgment from some of the world's leading number theorists.
What comes next
The process of verification could take years, and the community has historically been cautious about accepting major results.
The Clay Institute's requirement of two years of scrutiny sets a minimum bar, but the actual timeline depends on how quickly experts can digest the argument and whether they find any gaps.
Mathematicians at institutions including Princeton, the Institute for Advanced Study, and the Clay Institute itself are expected to begin reviewing the proof.
Given the complexity of the argument, initial assessments could take several months even with the Lean formalization providing a structural shortcut.
If verified, the proof would not only resolve one of mathematics's most famous open problems but also cement the role of AI language models as genuine research collaborators.
The Lean formalization means that the structural logic can be checked programmatically, reducing the risk of hidden errors that have plagued human-only proofs of major results in the past.
But the deeper question of whether the conceptual framing is correct, and whether the argument genuinely establishes the Riemann hypothesis under standard mathematical foundations, will require judgment from domain experts rather than automated tools.
Alpoge's closing words captured the mood: "In a time of revolution in mathematics, I hope the argument provides a glimpse towards the miracles that lie ahead."
Whether the proof holds or not, the race to crack the hardest problems in mathematics with AI has clearly entered a new chapter.
Common questions
Has the Riemann hypothesis been solved?
Anthropic mathematician Levent Alpoge claims to have proven it with Claude, but the proof has not been independently verified.
The Clay Mathematics Institute requires that a solution withstand two years of scrutiny before the one million dollar Millennium Prize can be awarded.
Was AI used to prove the Riemann hypothesis?
Alpoge says Claude was integral to developing and formalizing the proof. The argument has been verified mechanically using the Lean proof checker, but the mathematical soundness of the human-AI collaboration has not yet been confirmed by independent experts.
Sources & further reading
How this story was made
Written by Simon Mustak with AI assistance and checked against the linked sources. Company performance claims are attributed to the company. Analysis reflects AiLookout’s interpretation; we have not independently tested the products discussed. Cover photography is illustrative and does not depict the specific announcement or product.
Our editorial standards





