Subvolts take AI needs to understand space, not just words, for its next big leap. Dec 23, 2025 - Developers
- Learn about the limitations of current LLMs and the potential of spatial AI for building more capable systems.
- Business
- Understand the next frontier in AI development and its potential to reshape robotics, creativity, and computational interfaces.
A leading AI researcher explains a fundamental shift needed in AI development beyond current language models.
Fei-Fei Li: World Models and the Multiverse
Audio streams from a16z. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Fei-Fei Li joins Martin Casado and Erik Torenberg on the AI + a16z podcast to discuss world models. Li, CEO of World Labs, explains why AI needs spatial intelligence to understand the physical world, not just text. The conversation explores how this could advance robotics, creativity, and computational interfaces beyond current large language model limitations.
Imagine teaching a computer to understand a room like a person does, not just describe it. Fei-Fei Li explains why this spatial understanding is crucial for AI to truly grasp the world and how it could lead to smarter robots and new creative tools.
This episode is for anyone interested in the future of AI beyond language. Li argues that true AI understanding requires spatial reasoning, a gap current models like those from OpenAI and Google Gemini do not fully address. The discussion highlights how World Labs aims to build AI that can interact with and understand the real world, potentially impacting robotics and creative tools. It connects to the frontier labs' ongoing work in multimodal AI and embodied intelligence.
- Hosta16z
- GuestFei-Fei Licofounder and CEO, World Labs
- Guest: Fei-Fei Li
- Title: Fei-Fei Li: World Models and the Multiverse
- Show: AI + a16z
- Length: 23 minutes
- Concept: World models: AI systems that understand and reason about the 3D, physical world.
What if the next leap in artificial intelligence isn’t about better language—but better understanding of space?
In this episode, a16z General Partner Erik Torenberg moderates a conversation with Fei-Fei Li, cofounder and CEO of World Labs, and a16z General Partner Martin Casado, an early investor in the company. Together, they dive into the concept of world models—AI systems that can understand and reason about the 3D, physical world, not just generate text.
Often called the “godmother of AI,” Fei-Fei explains why spatial intelligence is a fundamental and still-missing piece of today’s AI—and why she’s building an entire company to solve it. Martin shares how he and Fei-Fei aligned on this vision long before it became fashionable, and why it could reshape the future of robotics, creativity, and computational interfaces.
From the limits of LLMs to the promise of embodied intelligence, this conversation blends personal stories with deep technical insights—exploring what it really means to build AI that understands the real (and virtual) world.
Follow Fei-Fei Li:
https://x.com/drfeifei
Follow Martin Casado:
https://x.com/martin_casado
Stay Updated:
If you enjoyed this episode, be sure to like, subscribe, and share with your friends!
Find a16z on X: https://x.com/a16z
Find a16z on LinkedIn: https://www.linkedin.com/company/a16z
Listen to the a16z Podcast on Spotify: https://open.spotify.com/show/5bC65RDvs3oxnLyqqvkUYX
Listen to the a16z Podcast on Apple Podcasts: https://podcasts.apple.com/us/podcast/a16z-podcast/id842818711
Follow our host: https://x.com/eriktorenberg
Check out everything a16z is doing with artificial intelligence here, including articles, projects, and more podcasts.
Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
- world models
- Fei-Fei Li
- AI systems that understand and reason about the 3D, physical world.
- spatial intelligence
- Fei-Fei Li
- AI's ability to understand and reason about space and physical objects.
- LLMs
- Martin Casado
- Large Language Models, AI systems that excel at generating and understanding text.
- embodied intelligence
- Martin Casado
- AI that can interact with and learn from the physical world through a body, like a robot.
24 minHow AI Is Rewriting the Rules of Cybersecurity | Truffle Security & Socket
WithDylan Ayrey, Feross Aboukhadijeh
47 minInside vLLM: The Engine Powering Open-Source AI
WithSimon Mo
41 minBuilding Self-Accelerating AI with Mirendil
WithBehnam Neyshabur, Harsh Mehta
43 minIdeogram’s Open-Weights Image Model and the Future of AI Design
WithMohammad Norouzi
Same guests, other shows
28 minHow Life Changes When We Reach Artificial Superintelligence w/ Dr. Fei-Fei Li & Dr. Eric Schmidt | EP #206
WithFei-Fei Li, Eric Schmidt
1 hFei-Fei Li: Human-Centered AI
WithFei-Fei Li
Questions people ask
- What are world models in AI?
- World models are AI systems designed to understand and reason about the 3D physical world, going beyond just generating text.
- Why is spatial intelligence important for AI?
- Spatial intelligence is considered a fundamental, yet missing, piece of current AI, necessary for true understanding and interaction with the real world.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to a16z.