Subvolts take NVIDIA exec explains how open models and world models drive AI's move into physical systems. Oct 1, 2026 - Developers
- Learn how open models and world models are shaping AI's expansion into robotics and autonomous systems.
- Business
- Understand NVIDIA's vision for physical AI and how open innovation drives progress in real-world AI applications.
A solid interview with an NVIDIA executive discussing the critical shift of AI into physical applications and the role of open models.
Open models and the future of Physical AI with NVIDIA
Audio streams from Daniel Whitenack and Chris Benson. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Daniel Whitenack and Chris Benson host Ming-Yu Liu of NVIDIA on Practical AI to discuss the future of physical AI. They explore how AI is moving beyond the cloud into robots and vehicles, emphasizing the importance of open models for innovation. The conversation covers world models for understanding the physical world and the development of increasingly capable physical AI systems.
AI is starting to work in the real world, like in robots and cars. This episode explains how making AI models open and teaching them to understand the physical world helps them get smarter and more useful for these tasks.
This episode offers insights into the growing field of physical AI, where machine learning interacts with the real world. It's for developers and researchers interested in robotics, autonomous systems, and the practical application of AI beyond screens. Liu's perspective from NVIDIA connects to frontier labs' work on multimodal models and embodied AI, highlighting how open models can accelerate progress in creating AI that can perceive, reason about, and act within physical environments.
- HostDaniel Whitenack
- HostChris Benson
- GuestMing-Yu LiuVice President of Cosmos Lab, NVIDIA
- Guest: Ming-Yu Liu, Vice President of Cosmos Lab at NVIDIA
- Show: Practical AI
- Length: 47 minutes
- Topic: Open models and the future of Physical AI
Chapters from the publisher ↗ · Transcript ↗ Key moments and transcript search come with a later piece.
As AI moves beyond the cloud and into robots, vehicles, and other physical systems, new approaches to models, simulation, and inference are emerging. Daniel and Chris are joined by Ming-Yu Liu, Vice President of Cosmos Lab at NVIDIA, who shares his perspective on open models, world models, and the future of physical AI. They explore why open models matter for innovation, how world models can help AI understand and simulate the physical world, and what it could take to build increasingly capable physical AI systems.
Featuring:
- Ming-Yu Liu – LinkedIn
- Daniel Whitenack – Website, GitHub, X
- Chris Benson – Website, LinkedIn, Bluesky, GitHub, X
Links:
Sponsors:
- Midwest AI Summit: Join AI practitioners on October 15 in Indianapolis for practical sessions, hands-on discussions, and real-world AI solutions. Use code PracticalAI20 to save 20% on your registration. https://midwestaisummit.com/#tickets
- Prediction Guard: A self-hosted AI control plane for running agents in high impact environments. predictionguard.com/practicalai
Resources and Events:
- physical AI
- Ming-Yu Liu
- AI that interacts with the real world, like in robots or cars.
- world models
- Ming-Yu Liu
- AI systems that learn to understand and simulate how the physical world works.
44 minNarrative intelligence and the human advantage
WithReed Frerichs
49 minFrom AGENTS.md to Enterprise Deployment
WithNick Kuhn
55 minHow to get discovered in AI search
WithLiam Dunne, Ben Moore
Same guests, other shows
26 minQuestions people ask
- What is the focus of the NVIDIA Cosmos Lab?
- The NVIDIA Cosmos Lab focuses on advancing AI's integration into physical systems like robots and vehicles, exploring new approaches to models, simulation, and inference.
- Why are open models important for physical AI?
- Open models are considered crucial for fostering innovation in physical AI, allowing broader access and collaboration to accelerate development.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Daniel Whitenack and Chris Benson.