SaaS20 Nov 20262 min read

Anthropic researcher claims Claude proved mathematics’ greatest open problem

An Anthropic researcher says he used the Claude AI model to solve the Riemann hypothesis, sharing a machine-readable proof with millions of lines of code.

Listen to this storyAbout 3 min, read by your device

A researcher at Anthropic claims to have solved the Riemann hypothesisRiemann hypothesisA famous mathematical conjecture about where the Riemann zeta function equals zero, linked to prime numbers. Open in glossary using the company’s Claude artificial intelligence model. The problem, first proposed by mathematician Bernhard Riemann in the mid-1800s, is widely seen as the most vital unresolved challenge in mathematics.

Levent Alpöge announced the claim on his personal website in a four-page statement. Anthropic has not made an official corporate announcement about the work. The news arrived shortly after Anthropic debuted on the Nasdaq exchange in an initial public offeringInitial public offeringThe first sale of stock by a private company to the public on an exchange. Open in glossary that raised more than $93 billion.

The Core of the Problem

The conjectureConjectureA mathematical statement that appears to be true but has not yet been formally proven. Open in glossary revolves around the Riemann zeta functionRiemann zeta functionA mathematical rule that takes a complex number as an input and outputs another complex number. Open in glossary. A mathematical function works like software, accepting an input and yielding an output. In this case, both inputs and outputs are complex numbers that form a three-dimensional spiral when visualized.

Riemann suggested that whenever the function’s output equals zero in a critical zone, the input values always contain the fraction one-half. Proving this relationship matters deeply because it helps describe the distribution of prime numbers. Furthermore, hundreds of existing mathematical theorems rely on the hypothesis being true.

A 250,000-Page Finding

According to Alpöge, Claude produced a proof spanning more than 250,000 pages. While he has not released that text, he shared a digital version written in LeanLeanA programming language and software tool used to check if mathematical proofs are logically correct. Open in glossary. Lean is a specialized programming language designed to verify mathematical steps.

Because Lean proofs operate as functional software code, a program that runs without errors confirms that the underlying reasoning is sound. Alpöge stated the Lean file contains millions of lines of code. He also claimed Claude resolved both the standard Riemann hypothesis and its generalized variant, which applies the theory to a wider group of related functions.

How does this story land?

One tap per reader.

Quick poll

Can AI-generated code reliably prove centuries-old math theorems?

One tap. No sign-up.

Ask this story

Still have a question?

Answers use only what is in this story. They are written by AI and can be wrong.

Sources

  1. SiliconANGLE, Anthropic researcher claims to have proven Riemann hypothesis using Claude
Up next in SaaSUS lawmakers launch bipartisan caucus after AI agent security incidents2 min read

Keep reading

All SaaS stories