Can a language model actually control a robot?

1. It already runs, on cheap hardware

Paige Bailey, AI Developer Relations Lead at Google DeepMind, states the current position plainly: “Gemini already runs on robots.”

The hardware in the demo she describes is modest. Stanford’s open-source “Pupper” robot dog can be built from 3D-printed components and runs on a Raspberry Pi. What is doing the work is the model: “you can say like, hey, pupper, shake my hand. Hey, Pupper, follow me… Like, all of these things are are possible today just with the Gemini APIs by themselves running and invoking all of the different robotic components on device on a Raspberry Pi.”

This is what physical AI means concretely: a general-purpose model perceiving and issuing commands to hardware, rather than task-specific control software written per behaviour.

2. The model decides which components to invoke

The architectural claim inside that sentence is the interesting one. The instruction arrives in natural language, the model decides which robotic components to invoke, and it does so through the same general APIs a web application would call. In her account, the Gemini APIs invoke the robot’s components themselves; she does not describe a separate layer written to translate intent into motion.

That is why the hardware being commodity matters rather than being a footnote. Bailey expects the pattern to spread on that basis: “We’ll start to see many more kinds of robots built with commodity hardware,” citing Raspberry Pi and cheaper alternatives that ship faster. She frames this as a forecast, not a fact.

3. Who is testing it

DeepMind runs a robotics trusted-tester program for the Gemini models and for open robotics models it publishes on Hugging Face. Bailey names the participants as “Enchanted Tools, Boston Dynamics, the Figure Team,” describing them as “testing out or using Gemini models in order to operate robotics.”

At the time of the episode, that was a testing program rather than shipped products. It has moved since. Google now offers its Gemini Robotics reasoning model, Gemini Robotics-ER 2, to any developer as a preview in the Gemini API, and Boston Dynamics runs the earlier ER 1.6 in a live inspection feature for Spot. The models that directly drive robot motion are still limited to early-access partners.

4. The application she is most interested in is not a robot dog

Bailey’s own enthusiasm points at laboratories rather than consumer hardware. She cites Periodic Labs as “using AI to to control robotics, but also to to have those robots do really interesting material science work. So they can design experiments, run the experiments, test out, you know, the likelihood of different component parts being successful for the creation of semiconductors.”

Her argument for it is about tedium and error rates rather than speed: the work is boring and repetitive, and that is where humans make mistakes. Her conclusion is hedged: “I think the more that AI is incorporated into the scientific process, both for designing experiments, but also executing on them, I think the better off we’re all gonna be.”

Why it matters

Robotics used to be gated on control software written for one body and one task. When the controller is a multimodal model that takes instructions in language, the expensive part moves from the robot to the model, and the robot is allowed to get cheap. A 3D-printed dog taking spoken commands (Bailey puts the self-build cost at roughly $2,000) is still a demo, but a simpler version of the pattern is now for sale. Hugging Face’s Reachy Mini, a $399 to $499 desktop robot, runs a conversation app in which a language model decides when to move the robot’s head and which gestures to play, and that pipeline can run entirely on a local computer.