In this episode, Professor Hannah Fry speaks with Jack Parker-Holder and Shlomi Fruchter about Genie 3, a general-purpose world model that can generate an unprecedented diversity of interactive environments. The conversation covers how this model's auto-regressive nature allows for the creation of consistent, explorable worlds from text or image prompts, and how its capabilities differ from video generation models like Veo. 

Further reading

• Genie 3
• SIMA, A generalist AI agent for 3D virtual environments
• SIMA, Google DeepMind: The Podcast
• xLand

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! 

This week's episode is a slight departure from our usual deep dives. Join Paige Bailey, DevRel lead, as she guides Hannah Fry though some of her favorite AI tools. Having spent years understanding the 'what' of these models, Hannah finally gets to experience the 'how’ – from generating prompts and 'vibe coding', to creating her own version of the infamous spaghetti meme.

Learn more and try the tools yourself:

• Gemini: https://gemini.google.com/
• Google Labs: https://labs.google/
• AI Studio: https://aistudio.google.com/
• Veo 3: https://deepmind.google/models/veo/
• Flow: https://labs.google/flow/

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 composition: Eleni Shaw
• Audio engineer: Richard Courtice
• Production Manager: Dan Lazard
• Studio Manager: 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! 

Further reading:

• Natural forests of the world: paper, data and benchmarks
• Forest loss drivers: paper, summary from WRI, and blog from GFW
• Forest loss drivers code: Google Earth Engine; at WRI; at GFW; or Zenodo. 
• Deep learning based remote sensing (open source): Jeo, GeeFlow
• Species mapping paper: Arxiv
• Google resources: Google Earth Engine. Agri with Google, wildlife cameras
• Perch: code, paper
• Perch x coral reefs: Blog
• Agile Modelling: paper, code
• DolphinGemma: blog

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! 

In this episode, host Hannah Fry is joined by Max Jaderberg and Rebecca Paul of Isomorphic Labs to explore the future of drug discovery in the age of AI. They discuss how new technology, particularly AlphaFold 3, is revolutionizing the field by predicting the structure of life’s molecules, paving the way for faster and more efficient drug discovery.

They dig into the immense complexities of designing new drugs: How do you find the right molecular key for the right biological lock? How can AI help scientists understand disease better and overcome challenges like drug toxicity? And what about the diseases that are currently considered “undruggable”? Finally, they explore the ultimate impact of this technology, from the future of personalised medicine to the ambitious goal of being able to eventually design treatments for all diseases.

Further reading: 

• AlphaFold 3: https://www.nature.com/articles/s41586-024-07487-w 
• AlphaFold Server: https://alphafoldserver.com/ 
• Isomorphic Labs: https://www.isomorphiclabs.com/ 
• AlphaFold 3 code and weights: https://github.com/google-deepmind/alphafold3

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 composition: Eleni Shaw 
• Audio engineer: Richard Courtice 
• Production Manager: Dan Lazard 
• Studio Manager: 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! 

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:

• Gemini Robotics
• Advances in robot dexterity
• Gemini Robotics x YouTube 

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! 

In this episode, Professor Hannah Fry interviews Joelle Barral, Senior Director of Research at Google DeepMind, about AI in healthcare. They discuss existing AI applications including image analysis for diabetic retinopathy and the expansion of diagnostic tools as a result of multi-modal models. The conversation highlights AI's potential to improve healthcare delivery, personalize treatment, expand access worldwide, and ultimately, bring back the joy of practicing medicine.

Further reading:

• Med-Gemini 
• AMIE

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! 

In this episode, Hannah is once again joined by Murray Shanahan, Professor of Cognitive Robotics at Imperial College London and Principal Scientist at Google DeepMind, for a philosophical deep dive on AI. They explore everything from consciousness and metacognition in animals, to symbolic AI and neural networks. Murray also shares insights into his involvement with the film 'Ex Machina' and discusses the idea of reasoning, anthropomorphism, and the future of AI.

Further reading / listening: 

• Toward the future, S1 Ep 7: https://youtu.be/yf31XT1G1RQ?si=6mAEsQhKwKPWk9oH

___

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 composition: Eleni Shaw
• Audio engineer: Richard Courtice
• Production Manager: Dan Lazard
• Video Director: Bernardo Resende
• Video Editor: Alex Baro Cayetano, Bilal Merhi
• Audio Engineer: Richard Courtice
• Camera and Lighting Operator: Robert Messere
• Production Coordination: Zoey Roberts, Sarah Ellen Morton
• Visual Identity and Design: Rob Ashley
• Commissioned by Google DeepMind

___

• Subscribe to our channel to watch every episode: https://www.youtube.com/@googledeepmind

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! 

In this episode of Google DeepMind: The Podcast, VP of Reinforcement Learning, David Silver, describes his vision for the future of AI, exploring the concept of the "era of experience" versus the current "era of human data". Using AlphaGo and AlphaZero as examples, he highlights how these systems surpassed human capabilities by engaging in reinforcement learning without prior human knowledge. This approach contrasts with large language models, which depend on human data and feedback. Silver emphasizes the need to explore this path to drive AI progress and achieve artificial superintelligence.

Timestamps 

• 00:00 Introduction
• 01:50 Era of experience
• 03:45 AlphaZero
• 10:19 Move 37
• 15:20 Reinforcement learning and human feedback
• 24:30 AlphaProof
• 29:50 Math Olympiads
• 35:00 Experience based methods
• 42:56 Hannah's reflections
• 44:00 Fan Hui joins

___

Thanks to everyone who made this possible, including but not limited to: 

• Presenter: Professor Hannah Fry
• Series Producer: Dan Hardoon
• Series Editor: Rami Tzabar
• Commissioner & Producer: Emma Yousif
• Music Composition: Eleni Shaw
• Audio Engineer: Richard Courtice
• Production Manager: Dan Lazard
• Video Director and Editor: Bernardo Resende
• Video Studio Production: Nicholas Duke
• 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! 

In our final episode for the year, we explore Project Astra, a research prototype exploring future capabilities of a universal AI assistant that can understand the world around you. Host Hannah Fry is joined by Greg Wayne, Director in Research at Google DeepMind. They discuss the inspiration behind the research prototype, its current strengths and limitations, as well as potential future use cases. Hannah even gets the chance to put Project Astra's multilingual skills to the test.

Further reading / listening:

• Gemini 2.0
• Project Astra 
• Decoding Google Gemini with Jeff Dean
• Gaming, Goats & General Intelligence with Frederic Besse

Thanks to everyone who made this possible, including but not limited to: 

• Presenter: Professor Hannah Fry
• Series Producer: Dan Hardoon
• Editor: Rami Tzabar, TellTale Studios
• Commissioner & Producer: Emma Yousif
• Music composition: Eleni Shaw
• Camera Director and Video Editor: Bernardo Resende
• Audio Engineer: Perry Rogantin
• Video Studio Production: Nicholas Duke
• Video Editor: Bilal Merhi
• Video Production Design: James Barton
• Visual Identity and Design: Eleanor Tomlinson
• 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! 

In this episode, Hannah is joined by Oriol Vinyals, VP of Drastic Research and Gemini co-lead. They discuss the evolution of agents from single-task models to more general-purpose models capable of broader applications, like Gemini. Vinyals guides Hannah through the two-step process behind multi modal models: pre-training (imitation learning) and post-training (reinforcement learning). They discuss the complexities of scaling and the importance of innovation in architecture and training processes. They close on a quick whirlwind tour of some of the new agentic capabilities recently released by Google DeepMind. 

Note: To see all of the full length demos, including unedited versions, and other videos related to Gemini 2.0 head to YouTube.

Future reading/watching: 

• Gemini 2.0 
• Decoding Google Gemini with Jeff Dean
• Gaming, Goats & General Intelligence with Frederic Besse

Thanks to everyone who made this possible, including but not limited to: 

Presenter: Professor Hannah Fry

Series Producer: Dan Hardoon

Editor: Rami Tzabar, TellTale Studios 

Commissioner & Producer: Emma Yousif

Music composition: Eleni Shaw

Camera Director and Video Editor: Bernardo Resende

Audio Engineer: Perry Rogantin

Video Studio Production: Nicholas Duke

Video Editor: Bilal Merhi

Video Production Design: James Barton

Visual Identity and Design: Eleanor Tomlinson

Commissioned by Google DeepMind

— 

Subscribe to our YouTube channel 

Find us on X

Follow us on Instagram

Add us on Linkedin

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!