World Models: AI Systems That Predict Visual States and Physics
Overview
Wired reports that world models, AI systems that predict the next visual state of the world and simulate physics, are being built by Niantic Spatial, Wayve, Tencent and Odyssey.
Wayve uses dashcam footage to recreate hazardous roads for testing. Odyssey CEO Oliver Cameron says a world model takes an image and predicts the next visual state, unlike a language model, which predicts the next word.
Yann LeCun's startup, Advanced Machines Intelligence, raised over a billion dollars, according to the report. The report's headline says world models will soon be inescapable, but the article itself gives no adoption figures to support that claim.
Written by AI from the articles below · updated Nov 8, 5:10 AM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
- Wired · AINewsWorld Models Will Soon Be Inescapable. Here’s What You Need to Know
AINiantic Spatial, Wayve, Tencent and Odyssey are building world models that use visual inputs to predict the next state of the world and simulate physics. Wayve uses dashcam footage to recreate hazardous roads for testing, and Yann LeCun's Advanced Machines Intelligence startup raised over a billion dollars. Odyssey CEO Oliver Cameron says a world model takes an image and predicts the next visual state, unlike a language model that predicts the next word.
Heat Trend
Not enough continuous observations to show a trend yet.