World model glossary
Browse a world model glossary with focused definitions for spatial intelligence, physical AI, and interactive world generation.
Visual glossaryGlossary tiles
The copy stays short so new visitors can build a mental map quickly.
An AI system that learns how an environment changes over time and in response to actions.
A world model is not just trying to make a pretty frame. It is trying to understand what can happen next in a world.The ability to understand objects, layouts, camera position, movement, and relationships in space.
This is the part that lets AI reason about where things are and how a scene is arranged.AI systems designed to operate in or reason about physical environments, often through robotics or autonomous systems.
Physical AI is where world models meet robots, vehicles, sensors, and real-world constraints.A foundation model trained for world simulation or physical AI workflows rather than only language or static media.
Think of it as a base model for simulating environments, motion, and possible futures.The generation of environments that can be navigated, changed, or acted on after they are created.
The key test is whether the generated world keeps making sense after a user moves through it or changes it.