28 August 2026

Anthropic releases safety framework for AI controlling lab equipment

• Anthropic published Model Hardware Standard, a ruleset letting AI agents operate microscopes, robots, and manufacturing equipment while enforcing safety constraints.
• The framework addresses risks like AI damaging physical systems or hurting people, which recent experiments have shown is possible when models are manipulated.
• Scientists and engineers can use the standard to specify which hardware actions AI models should avoid, similar to how Anthropic's earlier protocol controlled software interactions.
• The company is working with manufacturers and will test with trusted partners before wider release, citing concern that AI agents could be misused for biological weapons or other harms.

How it was covered

The AlgorithmMIT Technology Review

Anthropic released an AI tool that lets agents autonomously operate microscopes, lasers, and robotic arms with safety rules for lab operations. This is Anthropic's first tool designed for the physical world.

Reported by Wired