Subvolts take Google DeepMind unveils Genie 3, an AI that builds interactive worlds from text or images. Aug 22, 2025 - Developers
- Learn about Genie 3's auto-regressive approach to generating interactive environments, potentially applicable to game development or simulation.
- Business
- Understand how AI is evolving to create complex virtual worlds, opening new possibilities for interactive content and simulations.
This episode details a significant advancement in AI's ability to generate complex, interactive environments, moving beyond simple media generation.
Genie 3: An Infinite World Model with Shlomi Fruchter and Jack Parker-Holder
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 podcast features Jack Parker-Holder and Shlomi Fruchter discussing Genie 3, a general-purpose world model. This model generates diverse interactive environments from text or image prompts using an auto-regressive approach. The conversation contrasts Genie 3's capabilities with video generation models like Veo, highlighting its ability to create consistent, explorable worlds.
This episode explains Genie 3, an AI that creates entire interactive worlds from simple text or picture instructions. It's like an AI that can build game levels or virtual spaces you can explore, offering a new way for AI to generate complex content.
This episode offers insight into Google DeepMind's advancements in generative AI, specifically world models. Listeners learn about Genie 3's unique auto-regressive generation process and its potential applications. It is for AI researchers, developers, and enthusiasts interested in the frontier of generative AI beyond simple image or video creation. This work aligns with the broader trend across labs like OpenAI, Meta, and Google Gemini to build more comprehensive and interactive AI systems.
- HostHannah Fry
- GuestJack Parker-HolderGoogle DeepMind
- GuestShlomi FruchterGoogle DeepMind
- Episode Title: Genie 3: An Infinite World Model with Shlomi Fruchter and Jack Parker-Holder
- Show: Google DeepMind: The Podcast
- Length: 57 minutes
- Model Discussed: Genie 3
- Model Capability: Generates interactive environments from text or image prompts
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!
Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
- auto-regressive
- Shlomi Fruchter and Jack Parker-Holder
- A method where the AI predicts the next part of a sequence based on what it has already generated.
- world model
- Shlomi Fruchter and Jack Parker-Holder
- An AI system that understands and can generate complex, interactive environments or scenarios.
44 minCan AI help us better predict the weather?
WithPeter Battaglia
45 minThe mathematics of AI uncertainty
WithZoubin Ghahramani
53 minSame guests, other shows
43 minProject Genie: Create and explore worlds
WithDiego Rivas, Shlomi Fruchter, Jack Parker-Holder
31 minBehind the scenes of Google's state-of-the-art "nano-banana" image model
WithNicole Brichtova, Kaushik Shivakumar, Mostafa Dehghani, Robert Riachi
Questions people ask
- What is Genie 3?
- Genie 3 is a general-purpose world model developed by Google DeepMind that can generate diverse, interactive environments from text or image prompts.
- How does Genie 3 differ from video generation models?
- Genie 3's auto-regressive nature allows it to create consistent, explorable worlds, distinguishing it from models like Veo which focus on video generation.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Hannah Fry.