Two groups verify claim that Claude helped solve Riemann hypothesis
What's the story
A mathematician at Anthropic has claimed that their AI software, Claude, helped solve the Riemann hypothesis. The claim was made by Turkish-American mathematician Levent Alpoge on his personal website. The Riemann hypothesis is one of the seven Millennium Prize Problems and had remained unsolved for nearly 170 years. Following Alpoge's announcement, two independent organizations confirmed his claim after examining the programming language he used to present the solution.
Verification process
Two organizations confirmed Alpoge's claim
The Lean Focused Research Organization and Mathlib Initiative both confirmed that Alpoge's proof of the Riemann hypothesis is correct. They congratulated him on this major achievement, marking a significant milestone in AI-assisted problem-solving.
Problem details
What is the Riemann hypothesis?
The Riemann hypothesis, proposed by German mathematician Bernhard Riemann in 1859, deals with prime numbers.
These are whole numbers greater than one that can only be divided by themselves and one.
Riemann questioned the frequency and distribution of prime numbers on the number line and developed a formula to estimate their quantity below a certain number.
AI assistance
Concerns among mathematicians over AI-assisted solutions
Alpoge's claim comes amid growing unease among mathematicians over private companies spending huge sums on AI to solve complex problems.
In September, OpenAI claimed that a non-public version of its AI tools had helped solve part of another Millennium Prize Problem.
However, the solution was achieved by employing 10,000 AI systems at a cost of $15 million, an approach that struck mathematicians as strange.