Google Announces Genie 4 for More Realistic Interactive AI Worlds

What changed

Google announced Genie 4, a world model for generating interactive AI experiences.

The update adds more realistic worlds, real-time changes through natural-language prompts, generated audio, and simulations assisted by Gemini 3.8 Flash. Experiences can now run for about two minutes, up from one minute with Genie 3.

Why it matters

The notable shift is not just longer sessions. Genie 4 moves the experience closer to something a user can shape while it is running: describe a change, and Google says the world can update in response.

That makes the model more useful for interactive prototypes and experiments, where audio, visuals and live edits need to work together. Doubling the reported duration also gives an experience more room to develop before it ends.

But the practical value is still unproven. The announcement does not establish how reliably the model follows prompts, when Genie 4 will be available, what it will cost, or how it performs outside Google’s own claims.