A mathematician at Anthropic, Levent Alpöge, is contending that the San Francisco-based artificial intelligence (AI) company’s software, Claude, has helped solve one of mathematics' most significant unsolved problems, the Riemann hypothesis.
Alpöge posted a note on his personal website on Friday, stating that Claude assisted in developing a solution to the problem over approximately six weeks. He also linked to computer code, written in Lean, which he maintains could verify the findings.
The Riemann hypothesis, proposed by German mathematician Bernhard Riemann in 1859, is a nearly 170-year-old problem concerning the distribution of prime numbers. It is one of the seven Millennium prize math problems, with a $1m reward offered by the Clay Mathematics Institute for its solution.
Mathematicians have begun working to confirm Alpöge’s proof. If confirmed, this could represent a further breakthrough in using AI for complex mathematical problems.