World Model
An internal representation that learns how an environment works and predicts the consequences of actions.
In plain language
World models let agents simulate outcomes before acting, with uses in autonomous driving, robotics and video generation. They need large environment datasets and suffer from compounding long-horizon errors. Unlike language models, they emphasize physical and spatiotemporal consistency.
Example
A self-driving system simulates the outcome of a lane change before executing it.
Recommended tools
Related terms
Artificial General Intelligence
General AI that matches or exceeds humans across a wide range of tasks and can transfer learning.
Embodied AI
AI that perceives, decides and acts through a physical body.
Text-to-Video
Generating video clips from text descriptions.
Agentic Workflow
Breaking a task into steps planned by AI, executed by tools and reviewed by humans.