Anthropic unveils Model Hardware Standard to let AI control physical devices
This digest was compiled by AI from multiple sources — links to the originals are below.
Anthropic introduced the Model Hardware Standard (MHS), a new protocol that lets AI models like Claude control physical devices and run experiments. The standard includes a tagging system for hardware constraints and safety limits, and is being tested with partners including AWS, Hugging Face, and Raspberry Pi. Anthropic plans to make MHS open source and agent-agnostic after the preview period.
Key Facts
- Anthropic's MHS lets AI models control physical devices, adjust instruments, and run experiments autonomously.
- MHS includes a standardized tagging system that encodes hardware constraints, adjustable parameters, and safety limits.
- Preview partners include Amazon Web Services (Strands Robots), Hugging Face (LeRobot), Raspberry Pi, Automata, and Universal Robots.
- Anthropic plans to make MHS open source and agent-agnostic after the preview period.
The Standard
Anthropic introduced the Model Hardware Standard (MHS) to enable AI models like Claude to control physical devices and run experiments. The standard includes a tagging system that encodes hardware constraints, adjustable parameters, measurement options, and safety limits. These tags can be integrated into a reference file, allowing an AI model to quickly understand a device it has never encountered before.
Partner Ecosystem
Anthropic is working with a first group of scientific research labs and advanced manufacturers during the MHS preview period. Partners include Amazon Web Services (Strands Robots), Hugging Face (LeRobot), Raspberry Pi, Automata, and Universal Robots. These partners will help Anthropic build safety evaluations and develop best practices for AI systems operating physical equipment.
Future Plans
Anthropic plans to make MHS an open source and agent-agnostic standard for integrating AI and physical systems. In early testing with scientific partners over the last year, Anthropic said MHS reduced the time it took to integrate devices and made it possible to iterate faster in experimental settings. Anthropic's Kemeny said in a promo video that faster hypothesis testing could condense a century of progress into a decade.
2 sources
Anthropic unveils Model Hardware Standard to let AI control physical devices
