Model Hardware Standard: AI operating physical equipment

Summary
Anthropic’s Alek Kemeny and HHMI Janelia Research Campus postdoctoral scientist Dr. Arco Bast introduce the Model Hardware Standard (MHS), an open interface standard designed to connect AI models directly to laboratory and physical instruments. The video highlights collaborative implementations with partners like Danaher, Genentech, and HHMI Janelia, illustrating how AI agents such as Claude can autonomously control equipment and run scientific experiments.

What is shown

• [00:00] Manual preparation of a specimen slide on a Leica microscope.
• [00:09] Title card: “Previewing the Model Hardware Standard”.
• [00:31] Architectural diagram of lab setup “Before MHS,” showing tangled, custom point-to-point software integrations across microscopes, control PCs, cameras, centrifuges, and sensors.
• [00:46] Architectural diagram of “After MHS,” illustrating an AI agent communicating through a single MHS interface linked to all laboratory hardware.
• [01:06] Danaher demonstration: Claude executing terminal commands to control a Leica microscope stage, focus, scan slides, detect bacteria, and select imaging targets.
• [01:15] Genentech demonstration: Footage of robotic liquid handlers and lab automation monitoring screens executing an experiment parsed from a PDF.
• [01:41] HHMI Janelia demonstration: Real-time neural imaging in brain tissue, showing Claude directing microscope navigation, depth adjustment, and angle capture.

Claims & numbers

• Arco Bast states that experiments that previously took weeks now take days with AI hardware integration [00:01].
• Bast claims that prior to MHS, developing custom software integrations for complicated multi-device experiments required weeks of work [00:43].
• Bast states that under MHS, devices communicate at bare-metal speed [00:59].
• Alek Kemeny claims that at Genentech, an experiment outlined in a PDF was autonomously executed by Claude, which successfully recovered from errors overnight [01:17].
• Kemeny states that accelerating scientific iteration through MHS can help compress “a century of progress... into a decade” [02:05].

Notable quotes

• “There’s no common way to connect a model to physical equipment. The Model Hardware Standard changes that.” — Alek Kemeny [00:19]
• “MHS gives any AI model one standard way to connect with and operate devices.” — Arco Bast, MD [00:25]
• “This is how a century of progress can compress into a decade.” — Alek Kemeny [02:05]

Assessment
This is an official promotional preview produced jointly by Anthropic and the HHMI Janelia Research Campus. While real workflow captures (terminal outputs, live microscopy, automated lab machinery) are displayed, the footage is presented as a polished highlight reel rather than an unbroken, end-to-end technical demonstration.

Described by gemini-3.8-flash on 2026-09-29 from the video’s audio and frames.