Stay up to date with notifications from The Independent

Notifications can be managed in browser preferences.

Google DeepMind eyes AI robots after hiring Boston Dynamics CTO

Google wants to build robots that perform tasks at a similar or better level than humans

Anthony Cuthbertson
Thursday 20 November 2025 16:55 GMT
Comments
Robotic devices from Boston Dynamics and Unitree are seen at the International Conference on Robotics and Automation (ICRA) at ExCel on 30 May, 2023 in London, England
Robotic devices from Boston Dynamics and Unitree are seen at the International Conference on Robotics and Automation (ICRA) at ExCel on 30 May, 2023 in London, England (Getty Images)

Google’s AI division DeepMind is pushing further into robotics after hiring a former executive from leading robot maker Boston Dynamics.

Aaron Saunders, who served as the chief technology officer at the robotics firm, oversaw the development of both four-legged and humanoid machines during his 22 years at Boston Dynamics.

He joins Google DeepMind as its vice president of hardware engineering, with the hire seen as a core part of integrating the AI model Gemini into robots.

The tech giant has already demonstrated how Gemini can give robots spatial understanding and reasoning abilities in order to carry out physical tasks in the real world.

Demonstrations of Gemini Robotics models have included robots folding origami, preparing a packed lunch, tying show laces, and even figuring out how to slam dunk a basketball.

Google DeepMind CEO Demis Hassabis has previously spoken of his ambition to create a sort of AI operating system for robots that can work on any machine – similar to how Google’s Android operating system currently runs on more than 3 billion devices worldwide.

This could see Google focus on building the "brain" of humanoid robots, allowing them to perform tasks at a similar or better level than humans.

Google DeepMind launched a new version of Gemini Robotics earlier this year that allows AI robots to operate without an internet connection.

The approach marks a major shift for AI robots, which typically rely on far-away servers to carry out the necessary computing in order to operate.

By having an on-device system, humanoid robots are able to operate in any location on Earth – or in space – without data connectivity.

"Gemini Robotics On-Device marks a step forward in making powerful robotics models more accessible and adaptable," Carolina Parada, DeepMind’s head of robotics, said at the time.

"We’re excited to see what the robotics community will build with these new tools as we continue to explore the future of bringing AI into the physical world."

Join our commenting forum

Join thought-provoking conversations, follow other Independent readers and see their replies

Comments

Thank you for registering

Please refresh the page or navigate to another page on the site to be automatically logged inPlease refresh your browser to be logged in

[フレーム]

AltStyle によって変換されたページ (->オリジナル) /