News

Google’s Genie 4 expands AI-generated worlds with audio and real-time control

Google DeepMind has announced Genie 4, an updated world model for generating interactive AI experiences. Compared with the previous Genie 3 model, Genie 4 can create worlds intended to feel more realistic, add audio, and be modified while it is running through natural-language prompts.

The model uses Gemini 3.8 Flash to support its simulations and can generate experiences lasting about two minutes, roughly twice the length supported by Genie 3. Google is presenting the system as a potential tool for filmmakers seeking pre-visualization, teams prototyping game ideas, and developers creating training scenarios for robots.

Genie 4 is available in research preview for trusted testers, according to Google.

Sources