Google has opened Genie 4, its most realistic interactive world model so far, to paying users. On 18 November, Google DeepMind and Google Labs began rolling out a new Project Genie experience to eligible Google AI Ultra subscribers, with Google AI Pro members able to join a waitlist.
Genie 4 does not make a video to watch. It builds a small world you can move through, like a video game generated on the fly, from a text prompt, an image or both. The model creates a continuous interactive world at 720p and 24 frames per second, with synchronised audio. Users can type a new prompt while they are inside the world — to make it rain, add a character or change the action — without restarting it.
The update also brings real-time native audio for the first time in Project Genie, from ambient soundscapes to character dialogue that adapts as the scene changes. A conversational agent helps build worlds, Gemini watches the screen and suggests where to go next, and a community gallery lets people share and remix each other’s worlds. Genie 3 has been upgraded too, with sessions of up to three minutes per generation, up from 60 seconds.
On 19 November, developers gained access to the Genie 4 model through the Gemini API and Google AI Studio. Genie 4 was first presented in October as a research preview for a limited group of trusted testers.
Google is direct that these remain research models. The company lists open challenges in camera latency, complex physical interactions, real-time audio precision and long-term scene permanence. All generated video and audio carries SynthID watermarking, and inputs and outputs go through safety screening.
The launch lands in a fast-consolidating field. TechCrunch confirmed AMD has acquired World Labs and Amazon is pursuing Decart, leaving no independent world-model players at scale.
What it means
World models are moving from demo clips to tools people can play with and remix, closer to a game engine than a video generator. For now Genie 4 sits behind Google’s most expensive subscription, so it is a test of whether interactive AI worlds can find a paying audience — and a new reason for creators and game studios to watch Google’s AI plans.
Sources:
- i10X News, “Genie 4 Now Available in Project Genie for AI Ultra”, 18 November 2026 (reporting Google’s blog post by Diego Rivas and Pavan Kumar) — https://i10x.ai/news/genie-4-project-genie-google-ai-ultra-available
- Asia Daily, “Google’s Project Genie Turns Text Prompts Into Interactive Worlds”, 20 November 2026 — https://asiadaily.org/news/19241/
- Google DeepMind, Genie 4 model page — https://deepmind.google/models/genie/

