AI agents develop physical model from quantum-material observations

Researchers report in an arXiv preprint that their AI Theorist system autonomously developed a physical model explaining previously unpublished experimental observations in α-RuCl₃, a candidate material for realizing a Kitaev quantum spin liquid. The system combines hypothesis generation, first-principles calculations and refinement against experimental evidence. Applied to optical spectra, it produced a new interpretation of optical and photocurrent observations, identifying distinct excitonic states with different optical selection rules and spatial distributions. The authors describe this as, to their knowledge, the first autonomous AI demonstration of this kind using electronic-structure and many-body calculations. The demonstration concerns one quantum material.