Radical AI's self-driving lab at Sandia Labs spins an AI flywheel for materials discovery

CatAstro_Piyush · x · 2026-11-14

AI-for-Science startup Radical AI has deployed its "AI materials flywheel" at Sandia National Laboratories to speed up energy materials discovery.

The bottleneck: AI can generate a glut of new material hypotheses, creating more candidates than labs can physically test.
The approach: A self-driving lab combines AI, robotics, and real physical experiments to rapidly screen the best material candidates; each sample run sharpens the system's knowledge and shapes the next experiment.
Sandia staff scientist Vitalie Stavila notes that simulation only gets you so far—experimental data is what truly accelerates discovery.

Co-founded two years ago by Joseph Krause, the system targets semiconductor and next-generation energy materials.

More from Embodied