What changed: Google introduced Genie 4, the successor to Genie 3 which launched about a year earlier. The new model generates more photorealistic and interactive worlds, lets users edit scenes mid-simulation with natural-language prompts, generates accompanying audio such as music, ambient sound and dialogue, and extends scene length from one minute to about two minutes. Genie 4 is currently available only as a research preview to trusted testers; Google has targeted bringing it to Google AI Ultra subscribers via Project Genie, and to early-access partners of the Gemini API and Google AI Studio, in the coming weeks, though this timeline is not final.
Who is affected
The source points to filmmakers doing pre-visualization, game developers prototyping ideas, and teams building training scenarios for robots; it does not name other types of businesses.
What it means for your business
For teams using Project Genie or Google AI Ultra, this means early access to a tool that can generate interactive scenes and audio from text prompts during ideation; the model, however, remains limited to trusted testers and is not generally available.
What to watch
No action is required right now. Apply to the Project Genie waitlist to try the model in a test setting, and wait for Google's targeted general rollout and clarity on limitations such as scene-length drift and audio accuracy before relying on it for production work.
Timing
The source gives no fixed date; Genie 4 is in research preview now, with wider Google AI Ultra access only targeted for 'the coming weeks' and not yet final.
Source (press reports): The Verge, TechCrunch
