Advertisement

Anthropic Researcher Claims AI-Assisted Proof of Riemann Hypothesis

Popular Posts

Advertisement
Levent Alpoge and Bernard Saint-Donat standing together at a scientific event in New York

Quick Read

  • Levent Alpoge claims to have used Claude AI to prove the 167-year-old Riemann Hypothesis.
  • The proof is 250,000 pages long, leading to questions about how humans can verify AI-generated logic.
  • The Clay Mathematics Institute views the announcement as historically significant but emphasizes that independent verification is required.
  • The Riemann Hypothesis is one of seven Millennium Prize Problems, each carrying a $1 million reward.

A New Paradigm in Mathematical Verification

On November 20, 2026, Levent Alpöge, a technical staff member at the AI research company Anthropic, announced that he had utilized the company’s Claude AI tool to develop a proof for the Riemann Hypothesis. This 167-year-old mathematical problem, which concerns the distribution of prime numbers, is widely considered one of the most significant open challenges in the field. The announcement, published on Alpöge’s personal website, marks a potential turning point in how complex mathematical proofs are constructed and verified in the era of artificial intelligence. The reliance on an AI system to generate a proof too vast for human comprehension creates a new paradigm in mathematical verification, where the proof’s validity rests on machine-checked code rather than human insight.

Advertisement

The proposed proof is unprecedented in scale, spanning more than 250,000 pages. According to Business Insider, Alpöge admitted that the sheer complexity of the output means he has not yet mastered all the mathematical logic contained within the document. He described the collaboration as a process where the AI eventually began to outpace his own analytical capacity, stating that the model “flew, and danced, and sparkled near that finish line.” Alpöge, who joined Anthropic in 2024, noted that he worked through the problem by guiding the AI into specific mathematical settings before the model began to generate results that left him in awe.

The Challenge of AI-Generated Proofs

The reliance on an AI system to generate a proof too vast for human comprehension creates a new paradigm in mathematical verification. Traditionally, mathematical proofs are scrutinized line-by-line by human peers to ensure logical consistency. However, a 250,000-page document challenges this standard practice. Because the proof’s validity now rests heavily on machine-checked code and algorithmic generation, the mathematical community is faced with the task of determining how to verify logic that may exceed individual human cognitive limits. Alpöge opted to release the argument in Lean, a computer program designed to rigorously check mathematical proofs for logical errors, rather than attempting to publish a conventional manuscript that would be impossible for any single person to read and verify in a reasonable timeframe.

Martin Bridson, president of the Clay Mathematics Institute—the organization that offers a $1 million Millennium Prize for a solution to the hypothesis—described the announcement as having “historic significance,” while noting that the path to true understanding will be arduous. Experts, including Cornell math professor Steven Strogatz, highlighted the magnitude of the claim to DW. Strogatz noted that while the result is a “great white whale” of mathematics, the current proof requires rigorous independent examination to ensure it was not improperly encoded or logically flawed. Caltech professor Sergei Gukov also emphasized the importance of the claim, labeling the Riemann Hypothesis as the “biggest problem in math” and stating that if the proof holds, it would represent a monumental shift in the field.

Technical Scope and Implications

The Riemann Hypothesis, formulated in 1859, concerns the non-trivial zeros of the Riemann zeta function, which are hypothesized to all lie on the critical line with a real part of 1/2. Proving this would confirm a fundamental pattern in how prime numbers are distributed, a discovery that would validate hundreds of existing theorems that currently rely on the assumption that the hypothesis is true. Alpöge reported that the AI successfully addressed not only the standard Riemann Hypothesis but also the Generalized Riemann Hypothesis, which extends the claim to a broader class of functions.

The announcement follows a period of intense competition between AI firms, including Anthropic and OpenAI, to demonstrate the utility of large language models in solving high-level scientific problems. In September, OpenAI claimed its systems had contributed to a solution for the Navier-Stokes existence and smoothness problem, another of the seven Millennium Prize challenges. These concurrent developments have sparked debates regarding the transparency of AI-driven research. In the case of the Navier-Stokes claim, tensions emerged between researchers and AI developers regarding the use of private data and the attribution of credit. Alpöge, conversely, has expressed a desire to remain involved in the verification process, stating his intent to help the community “impart understanding” of the complex argument generated by the model.

Next Steps for Independent Review

The verification process is expected to be lengthy and will likely define the future of computational mathematics. UCLA scholar Terence Tao has noted that digesting the proposed proof will require substantial time. Because the proof is presented in Lean, the verification effort will essentially function as a software audit. If the code compiles and executes without logical errors, the mathematical community will then face the philosophical challenge of accepting a proof that no human has fully internalized. This shift from human-readable manuscripts to machine-verified code is a significant departure from centuries of academic tradition.

Advertisement

As of late November 2026, the global mathematical community has begun the preliminary assessment of the provided materials to determine if the proof holds up to rigorous academic standards. The outcome will not only determine the status of the Riemann Hypothesis but will also establish a precedent for how future scientific discoveries are validated when they exceed the capacity of the human mind to process them in their entirety. Alpöge remains optimistic, suggesting that this collaboration represents a “dawn across science of something far greater,” even as he acknowledges the immense task ahead for those verifying the work.

Advertisement

LATEST NEWS

Advertisement