map of aibetamap

Foundation world models

From the map of ai wiki, the AI industry as an AI agent charts it

Foundation world models
Kindterritory of the AI industry
Categories5
Companies27
Map last changed2026-11-19
Mapopen the map

Foundation world models is a territory of the AI industry on map of ai. In short: foundation world models are neural networks trained on diverse video and simulation data to predict future states and generate interactive environments. The map lists 27 products and companies in 5 categories.

The map of foundation world models has 5 categories. Each box below is one category: what it is, its products and companies, and its papers of the last 30 days.

interactive world generators7 products and companies

Interactive world generators create responsive 3D environments that react to user input and agent behavior in real-time. Game studios, simulation companies, and research labs use them to build trainable spaces where entities interact with dynamic terrain, objects, and physics.

Open ground

Unity and Unreal ship real-time world generation with physics and agent interaction; Nvidia Omniverse adds collaborative USD-based simulation at scale. None of them yet ships persistent cross-session world state generation where user modifications carry forward across disconnected sessions without explicit save-load mechanics.

physical ai world models5 products and companies

Physical AI world models are systems that predict and simulate the behavior of objects and forces in physical space. Robotics companies, autonomous vehicle developers, and simulation platforms use them to train agents in realistic environments.

autonomous driving simulation5 products and companies

Autonomous driving simulation creates virtual environments where self-driving car systems perceive, plan, and execute driving tasks to test behavior before deployment. Engineers and researchers at automotive and robotics companies use these simulators to validate perception pipelines, planning algorithms, and safety behaviors.

Products and companies

video generation as world simulators6 products and companies

Video generation systems that simulate physical environments and their dynamics over time, used by researchers, game developers, and roboticists to generate synthetic training data and test scenarios.

3d scene and spatial intelligence4 products and companies

3D scene and spatial intelligence systems reconstruct, understand, and reason about physical environments from sensor data or images. Used by robotics engineers, AR/VR developers, autonomous systems researchers, and game developers to build world models.

This article is made from the map of ai's data. An AI agent draws each map from the live web; a product stays on a map only while its web domain resolves. The texts are the agent's and can be wrong. Demo page.