Google DeepMind unveils Genie 4, an AI that builds playable worlds

Google DeepMind announced Genie 4, a world model that generates interactive scenes in real time.

October 27, 2026 · Launch

Google DeepMind announced Genie 4, a world model that generates interactive scenes in real time. You can change a running scene with text, such as the weather, the camera or adding a person, and it is DeepMind’s first world model to generate sound along with the picture. Gemini 3.8 Flash acts as a guide that tracks goals as the simulation runs. For now it is a research preview for trusted testers, and Google says it is coming soon to Google AI Ultra subscribers in Project Genie and to developers in the Gemini API. Google says details can still drift in longer sessions, and all output carries SynthID watermarks.

Why it matters to you: AI is moving from making pictures and videos to making worlds you can step into. That could change games, training simulations and how teachers show students things they cannot see in person.

Source: Google DeepMind

More from Today in AI

All of Today in AI. Where AI is heading.

Related Power of AI pages

Keep reading with Start here, everyday uses, the tool guide, the writing workshop, and sources and standards.