Subvolts take Robots are getting smarter and more dexterous, learning complex tasks and adapting to new situations. May 22, 2025 - Developers
- Learn about the latest breakthroughs in robot control and learning, enabling more general physical AI.
- Business
- Understand how advanced robotics can lead to new applications and efficiencies in physical tasks and automation.
This episode features a leader from a top AI lab discussing significant progress in robotics, a key area of AI development.
Redefining Robotics with Carolina Parada
Audio streams from Hannah Fry. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Google DeepMind's Carolina Parada joins Hannah Fry to discuss recent robotics advancements. They cover multimodal understanding and embodied reasoning, enabling robots to interact with the physical world more generally. The episode explores a two-system approach for complex reasoning and rapid reactions, showcasing robots learning dexterous tasks and adapting to new situations in real-time.
Robots are learning to understand and interact with the world like humans, using both slow thinking for planning and fast reactions. This allows them to do tricky tasks like tying shoelaces and quickly adapt to unexpected changes.
This episode offers insight into the cutting edge of robotics from a Google DeepMind leader. Listeners learn how robots are gaining more general capabilities through multimodal understanding and embodied reasoning. It is for anyone interested in AI's physical applications, from researchers to tech enthusiasts. This directly relates to current frontier lab efforts in creating more capable and adaptable AI systems that can interact with the real world.
- HostHannah Fry
- GuestCarolina ParadaSenior Director and Head of Robotics, Google DeepMind
- Guest: Carolina Parada, Senior Director and Head of Robotics at Google DeepMind
- Show: Google DeepMind: The Podcast
- Length: 45 minutes
- Key Concepts: Multimodal understanding, embodied reasoning, two-system approach ('slow and fast thinking')
- Example Tasks: Tying shoelaces, adapting to new scenarios in real-time
In this episode, Hannah is joined by Carolina Parada, Senior Director and Head of Robotics at Google DeepMind. They explore the recent leap forward in robotic capabilities, highlighting advancements in multimodal understanding and embodied reasoning, which enable robots to interact with the physical world with unprecedented generality.
They dig into the two-system approach - ‘slow and fast thinking‘ - that enables both complex reasoning and rapid, reactive movements. Using examples from robots learning dexterous tasks such as tying shoelaces to adapting to entirely new scenarios in real-time, Parada highlights how key breakthroughs in understanding, dexterity, and control are now coming together to rapidly advance robotics unlike ever before.
Further reading/viewing:
Thanks to everyone who made this possible, including but not limited to:
- Presenter: Professor Hannah Fry
- Series Producer: Dan Hardoon
- Editor: Rami Tzabar
- Commissioner & Producer: Emma Yousif
- Music: Eleni Shaw
- Audio Engineer: Richard Courtice
- Production Manager: Dan Lazard
- Video Studio Production: Nicholas Duke Video
- Director: Bernardo Resende
- Video Editor: Bilal Merhi
- Audio Engineer: Perry Rogantin
- Camera and Lighting Operator: Robert Messere
- Production Coordination: Zoey Roberts, Sarah Ellen Morton
- Visual Identity and Design: Rob Ashley
- Commissioned by Google DeepMind
Please leave us a review on Spotify or Apple Podcasts if you enjoyed this episode. We always want to hear from our audience whether that's in the form of feedback, new idea or a guest recommendation!
Please leave us a review on Spotify or Apple Podcasts if you enjoyed this episode. We always want to hear from our audience whether that's in the form of feedback, new idea or a guest recommendation!
Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
- Multimodal understanding
- Carolina Parada
- Robots understanding information from different sources like text, images, and touch.
- Embodied reasoning
- Carolina Parada
- Robots using their physical body and senses to think and solve problems in the real world.
44 minCan AI help us better predict the weather?
WithPeter Battaglia
45 minThe mathematics of AI uncertainty
WithZoubin Ghahramani
53 minSame guests, other shows
39 minIntroducing Gemini Robotics 2
WithCarolina Parada, Stuart Bowers, Kanishka Rao, Jie Tan
40 minGoogle I/O 2025 Recap with Josh Woodward and Tulsee Doshi
WithJosh Woodward, Tulsee Doshi
Questions people ask
- What is the two-system approach in robotics?
- It combines 'slow and fast thinking' to enable both complex reasoning and rapid, reactive movements in robots.
- What kind of tasks can these new robots perform?
- They can learn dexterous tasks like tying shoelaces and adapt to entirely new scenarios in real-time.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Hannah Fry.