Agents in Science

Preprint: no AI agent setup passes more than 46 of 168 physics research runs

Researchers report FrontierPhysics, 56 tasks written by 42 physicists from their own work. Across 22 models in 26 configurations, none passed more than 46 of 168 rollouts; weak rigor and expert intuition were the main gaps.

Why it matters: Agents that handle physics questions still fall short on open-ended research. Use them for scoped sub-tasks and keep experts judging whether results hold up.