Unreleased Claude AI Model Makes Progress on Riemann Hypothesis-Related Problem
Anthropic announced that an unreleased research version of its Claude AI model has successfully increased the lower bound for the fraction of zeros of the Riemann zeta function that satisfy the Riemann hypothesis. While it did not solve the hypothesis itself, this represents a notable mathematical advancement achieved by an AI. This achievement demonstrates the growing capability of large language models (LLMs) to contribute to frontier mathematics and solve complex, unsolved theoretical problems. It highlights a shift from AI acting merely as a coding assistant to becoming an active collaborator in scientific and mathematical discovery. The AI focused on the critical line of the Riemann zeta function, improving the proven lower bound of nontrivial zeros that lie on this line. Specific numerical details of the new lower bound were not fully disclosed in the initial announcement.
## BACKGROUND
The Riemann hypothesis is one of the most famous unsolved problems in mathematics and one of the Millennium Prize Problems. It posits that all non-trivial zeros of the Riemann zeta function have a real part of 1/2, meaning they lie on a specific vertical line called the critical line. Over the decades, mathematicians have attempted to prove this by showing that at least a certain percentage (or lower bound) of these zeros must lie on this line.