Anthropic previewed a new standard called MHS that lets AI agents control lab machines like microscopes. It replaces incompatible device-specific APIs with one unified interface.
The standard was developed with HHMI and is now available to limited partners.
Anthropic plans to open-source it and expand to manufacturing equipment.
What happened
Anthropic PBC previewed the Model Hardware Standard (MHS), a unified interface that lets AI agents control scientific instruments such as microscopes. It was developed with HHMI and is currently available only to a limited number of partners.
Why it matters
MHS replaces device-specific APIs, which are often incompatible, with a single set of configuration commands. This makes it easier for researchers to automate lab workflows and allows AI agents to manage machines, as shown by HHMI's AI agent that operates microscopes and QuEra Computing's use of MHS with Claude to coordinate lasers.
What to watch
Anthropic plans to release MHS under an open-source license and says it could also control manufacturing equipment. Amazon Web Services, Hugging Face, and several industrial robot suppliers are already working to integrate MHS, while Anthropic will develop safety features with early adopters before broad availability.
Ask the AI about this article →
The Model Hardware Standard addresses a practical bottleneck in laboratory automation: the need to write custom code for each instrument's unique API. By offering a single configuration interface, MHS simplifies the programming task and, crucially, makes it feasible for AI agents to take over. Anthropic's internal test, where Claude generated a script to adjust laser positions, demonstrates the potential for cost savings compared to inference-based workflows.
The collaboration with HHMI, a medical research institute, and the early involvement of QuEra Computing and AWS suggest a focus on both scientific and industrial applications. The open-source plan could accelerate adoption, but safety features are still being developed, indicating that the standard is not yet ready for widespread use. The integration efforts by cloud and robotics companies may signal broader implications for manufacturing, though this remains a stated ambition rather than a completed outcome.
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytime
Ask AI anything about this article. Q&As are published on this page for other readers too.
Honor has moved its Robot Phone from an internal debate to a mass-produced commercial product, more than a yea…
Google announced Gemini Omni 1.1 Flash on August 27, a multimodal AI model for video generation and editing, a…
A survey by Japanese research firm Parol General Research Institute found that among regular employees, AI usa…
ZEALS and Timee announced a partnership on August 27, 2026, to create a new work category called "data collect…
Nvidia has taken a minority stake in Cloverleaf Infrastructure, a developer that secures powered land and util…
Russian-speaking hackers used SpaceX's Cursor AI agent to breach a Belgian chemical company and six other firm…
