Artificial-intelligence company Anthropic has introduced a research preview of the Model Hardware Standard, or MHS—a framework intended to let AI agents communicate with and operate programmable physical equipment.
The system could allow AI agents to control several laboratory or manufacturing devices simultaneously, including microscopes, robotic arms and liquid-handling machines. Anthropic says possible applications range from conducting routine drug-discovery experiments to calibrating lasers used in quantum computers.According to the company, the standard can work with any device equipped with a programmable interface. It is also designed to allow AI agents and machines to communicate across computer networks.
Anthropic is initially sharing the framework with selected scientific laboratories and advanced manufacturers. Safety evaluations will be conducted before the company considers releasing it as open-source technology.The development could eventually allow laboratories and production facilities to operate continuously with less direct human involvement. However, strong supervision and safety safeguards would be necessary before AI systems are permitted to control sensitive or potentially hazardous equipment independently.
