Anthropic's Model Hardware Standard lets AI agents run labs and factories
Previewing the Model Hardware Standard
Anthropic has opened a research preview of the Model Hardware Standard (MHS), a specification that lets AI agents operate physical devices like microscopes, liquid handlers, and robotic arms. Developed with HHMI Janelia, MHS reduces hardware integration from weeks to minutes, enabling autonomous experiments and real-time adjustments. Partners like Genentech, UW, and QuEra report faster workflows and improved reliability, such as a 99.3% laser lock recovery rate. MHS is model-agnostic and works with any programmable device, with plans to open-source it later.
The agent developed a controller that recovers the laser’s “lock”—the ultra-precise frequency the lasers must hold to interact with the atoms—99.3% of the time without human intervention.