Anthropic Previews Model Hardware Standard, Letting AI Agents Run Lab Instruments

Anthropic has opened a research preview of the Model Hardware Standard, a shared spec that lets AI agents operate microscopes, robotic arms and liquid handlers through one interface with built-in safety limits. It is model-agnostic and cuts hardware integration from weeks to hours. In tests at QuEra, laser recovery rose from 58% to 99.3%. Open-sourcing is planned after safety work.