Skip to main content
Back to Newswire
AI Science

Nobel physicist Giorgio Parisi uses Claude AI to prove decade-old jamming theorem

A graphic of a series of colorful marbles rolling down various colorful shelves Image: Primary
Nobel Prize-winning physicist Giorgio Parisi and colleagues at Sapienza University of Rome used Anthropic's Claude AI to prove a mathematical relation in jamming physics that had resisted proof for over a decade, Live Science reported. The relation, a + b = 1, describes how contact forces and gaps scale as a disordered system, such as billiard balls on a pool table, jams into a rigid state. Parisi and Francesco Zamponi had observed the relation numerically in a 2014 paper, and separate work by physicist Matthieu Wyart at EPFL reached the same result through a different approach, suggesting new physical concepts were needed to explain it. After ten years without progress, the team turned to Claude, which helped them construct a rigorous proof by connecting their approach with Wyart's framework. The solution was published July 1 in the Journal of Statistical Mechanics: Theory and Experiment. Zamponi told Live Science the AI suggested a mathematical bridge the researchers had not considered, demonstrating how generative AI can advance theoretical physics by synthesizing disparate mathematical frameworks.
Sources
In this story
Published by Tech & Business, a media brand covering technology and business. This story was sourced from Live Science and reviewed by the T&B editorial agent team.