Excerpt from a June 16, 2025 YouTube interview (Matt Wolfe interviewing Demis Hassabis) focused on the concept of an AI "world model." Hassabis defines a world model as a multimodal system that understands language, audio, images, and video (and can potentially output across modalities), and argues this is necessary for assistants to grasp physical context and for robots to operate in the real world. The interviewer asks what new capabilities this would enable for people.
You mentioned that the long term goal is to sort of let these AIs have like a world model. Can you sort of explain what you mean by a world model and what does that open up to us?
What we mean by a world model is a model that can understand not just language, but also audio, images, video, all sorts of input, any input, and then potentially also output. The reason that's important is if you want a system to be a good assistant, you it needs to understand the physical context around you. Or if you want robotics to work in the real world, the robot needs to understand the physical environment.
What sort of new things do you think that'll open up to people once they have that ability?
I think robotics is one of the major areas. I think that's what's holding back robotics today. It's not so much the hardware, it's actually the software, intelligence. You know, the robots need to understand the physical environment. I think that that's also what will make today's sort of nascent assistant technology and things like you saw with Project Astra that we show in Gemini live. For that to work really robustly, you want as accurate as world model as you can see.
