The model generates interactive scenes with native audio that users can change mid-simulation with text prompts; it is with trusted testers for now, with Google AI Ultra subscribers and some developers next.
Google DeepMind on Tuesday announced Genie 4, a world model that generates interactive environments in real time and, for the first time in the series, produces audio such as ambient sound, music and dialogue alongside the visuals. Users can change a running simulation with text prompts, for example to alter the weather, move the camera or add an object, while a Gemini model plans steps and checks frames to keep the scene coherent. Scenes can stay consistent for about two minutes, double Genie 3’s limit, according to TechCrunch and The Verge. Genie 4 is available as a research preview to trusted testers and is coming to Google AI Ultra subscribers in Project Genie and to early access partners of the Gemini API and Google AI Studio. DeepMind says details and objects can still drift in longer sessions, and that audio accuracy and legible in-world text remain unsolved. Generated video and audio carry its SynthID watermark.
Sources: Google DeepMind · TechCrunch · The Verge