Subvolts take Google's Project Genie prototype lets AI create and explore interactive worlds from simple prompts. Jan 30, 2026 - Developers
- Learn about new AI capabilities for generating interactive simulations and environments.
- Business
- Understand emerging AI research that could power future interactive applications and robotics.
This episode features key researchers discussing a significant AI research prototype for creating interactive worlds.
Project Genie: Create and explore worlds
Audio streams from Google AI. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Google AI: Release Notes features the Project Genie team discussing their experimental research prototype for world models. Hosts Logan Kilpatrick, Diego Rivas, Shlomi Fruchter, and Jack Parker-Holder explore how Genie creates and allows exploration of interactive worlds. The episode covers the technology's capabilities, potential applications, and future development.
Imagine teaching a computer to build and play in its own digital worlds, like a sandbox. Project Genie is a Google experiment that lets AI generate these worlds from basic instructions, allowing exploration and interaction. This could lead to more advanced AI that understands and manipulates complex environments.
This episode offers a look into Google DeepMind's "world models," a significant area of AI research. Listeners learn about creating interactive simulated environments, potentially for robotics and general AI advancement. It's for researchers and AI enthusiasts interested in how AI can understand and generate complex, dynamic systems. This work aligns with frontier labs' push towards more general AI capabilities beyond specific tasks.
- HostLogan Kilpatrick
- GuestDiego RivasGoogle DeepMind
- GuestShlomi FruchterGoogle DeepMind
- GuestJack Parker-HolderGoogle DeepMind
- Episode Title: Project Genie: Create and explore worlds
- Show: Google AI: Release Notes
- Length: 43 minutes
- Guests: Diego Rivas, Shlomi Fruchter, and Jack Parker-Holder from the Project Genie team
- Product: Project Genie is an experimental research prototype
Chapters:
00:00 - Intro and defining world models and RL roots
01:51 - Demo: Goldfish and shark in underwater world
04:59 - Project Genie gallery
06:31 - Physics, remixing, and UI prompts
11:00 - Demo: Nano Banana mascot “Bob”
13:20 - Constraints, generation limits, and infrastructure
17:04 - Trusted testers and robotics future
28:34 - Frontier prompting and universal simulation
29:27 - Cross-Google collaboration
31:16 - Adoption timelines and impact
34:16 - Model generalization and historical context
38:52 - Hardware limits and the slope of progress
- world models
- AI researchers
- AI systems that learn to understand and simulate how environments work.
- RL roots
- AI researchers
- Connections to Reinforcement Learning, a type of AI training.
27 minKoray Kavukcuoglu on frontier models, coding agents, and building AGI
WithKoray Kavukcuoglu
39 minIntroducing Gemini Robotics 2
WithCarolina Parada, Stuart Bowers, Kanishka Rao, Jie Tan
42 minGemini co-leads on project origins and what's next
WithJeff Dean, Koray Kavukcuoglu, Noam Shazeer, Oriol Vinyals
40 minSame guests, other shows
57 minGenie 3: An Infinite World Model with Shlomi Fruchter and Jack Parker-Holder
WithJack Parker-Holder, Shlomi Fruchter
2 h 14AGI Debate: Is It Finally Here? | EP #227
WithSalim Ismail, Dave Blundin, Alexander Wissner-Gross
2 h 50
51 minQuestions people ask
- What is Project Genie?
- Project Genie is an experimental research prototype from Google DeepMind that allows for the creation and exploration of interactive worlds using AI.
- What are world models in AI?
- World models are AI systems that learn to understand and simulate the dynamics of an environment, enabling them to predict outcomes and generate new scenarios.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Google AI.