Anthropic employee claims AI solved 150-year-old Riemann Hypothesis
A mathematician using Anthropic's Claude AI claims to have produced a 17-million-line proof for one of the world's most famous unsolved math problems.

The short version
- The Riemann Hypothesis has remained unsolved since it was first proposed in 1859.
- The purported proof consists of 17 million lines of Lean code, a language used for formal verification.
- The discovery was announced shortly after Anthropic's record-breaking $81 billion stock market debut.
- The Clay Mathematics Institute has not yet confirmed the solution, noting that a formal assessment is required.
Why it matters
If verified, this would be one of the most significant mathematical breakthroughs in history, potentially changing how we understand prime numbers. It also signals a shift toward using AI to handle massive, complex formal verifications that exceed human capacity.
The Riemann Hypothesis claim
Levent Alpöge, a member of the technical staff at Anthropic, announced on Friday that he believes he has solved the Riemann Hypothesis. This mathematical problem has remained unsolved for more than 150 years since it was first proposed in 1859. The hypothesis specifically concerns the distribution of prime numbers as they approach infinity.
Alpöge stated that he worked with Anthropic's AI tool, Claude, to reach this conclusion. He noted that while he and the AI worked together extensively, the tool eventually
The mechanics of formal verification
The proof was developed using Lean, which is a programming language designed for the formal verification of mathematical theorems. Unlike a standard written paper, this code allows a computer to check every logical step of an argument to ensure it is correct. Because computers are extremely precise, they can identify tiny errors that a human might overlook during a manual review.
Alex Kontorovich, a professor at Rutgers University, noted that the computer is much more picky than a human reviewer. He suggested that if a proof can satisfy the requirements of the computer, it is very likely to satisfy a human mathematician as well. However, he emphasized that a human still needs to do the work of reading and understanding the underlying logic to truly grasp the truth of the solution.
A pattern of AI-assisted breakthroughs
This claim follows a similar trend in the field of high-level mathematics and physics. In September, OpenAI reported that its technology played a significant role in a breakthrough involving the Navier-Stokes equation. That specific equation is used to describe how fluids move, and its solution is another major milestone in the scientific community.
Kontorovich described the rapid succession of these AI-driven milestones as an "out-of-body experience." He expressed a sense of disbelief at the speed at which these complex problems are being addressed by artificial intelligence. The transition from solving fluid dynamics to tackling the Riemann Hypothesis suggests a growing trend of AI being used to push the boundaries of human knowledge.
The road to validation and risk
Alpöge admitted that he is presenting a complete argument even though he has not yet mastered the massive neighborhood of mathematics surrounding it. He noted that his work with the AI produced more than 250,000 pages of content. Because of the sheer scale of this output, the journey from the current announcement to a place of real human understanding will be long and difficult.
The announcement comes at a time of significant growth and scrutiny for Anthropic. While the company is seeing major advances in scientific research, its leadership has also warned about the potential risks posed by artificial intelligence. As the company seeks to balance these breakthroughs with safety, the mathematical community will need to conduct a lengthy assessment to verify the validity of the new proof.
The stakes of the Millennium Prize
The Riemann Hypothesis is not just a theoretical puzzle; it is one of the most prestigious challenges in modern science. In 2000, the Clay Mathematics Institute designated it as one of seven Millennium Prize Problems. The institute offered a $1 million prize for the first person to provide a valid solution to each of these problems.
Martin Bridson, the president of the institute, described the announcement as having historic significance. He stated that the discovery holds the promise of a dramatic transformation in how people perceive what is possible when humans struggle to extend their understanding of mathematics.
Corporate silence and academic hurdles
Despite the magnitude of the claim, Anthropic has not yet provided an official statement on the matter. The company did not respond to an email from The Associated Press seeking a comment on the employee's purported discovery. This lack of response leaves the company's official stance on the AI's role in the proof unknown for now.
Academic experts emphasize that a computer-verified proof is only one part of the process. Professor Kontorovich noted that once a person is done with the code, they should have a deep understanding of why something is true. He stressed that the human researcher must still do the work to fully comprehend the logic behind the thousands of pages of output.
Sources and links
Read the original at United News of Bangladesh
First reported by United News of Bangladesh. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.
Liked it? Say why
Add a line for other readers.
No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.