Person
Fei-Fei Li
- Google DeepMind releases Genie 4 world model with real-time audio and on-the-fly scene changes
Google DeepMind on Tuesday, Oct 27, released its latest world model, Genie 4, as a research preview. It generates more realistic and interactive virtual worlds than its predecessor and can create matching music, ambient sound and dialogue in real time; users can type prompts during a simulation to change the weather or camera angle or add new characters and objects. Scenes stay consistent for up to two minutes, twice as long as Genie 3, with more realistic physics and people, and the model draws on Google's updated Gemini models. DeepMind sees world models as key to artificial general intelligence and as a source of unlimited training environments for physical robots; the industry is also consolidating, with AMD acquiring World Labs, founded by Fei-Fei Li.
Oct 28, 2026 Morning