Google’s Project Genie lets you create your own 3D interactive worlds

Last summer, Google DeepMind launched Genie 3. It’s a so-called global world, an AI system that can generate images and respond as the user moves through the environment simulated by the software. At the time, DeepMind positioned Genie 3 as an AI agent training tool. Now the model is being made available to people outside of Google to try with Project Genie.

To get started, you’ll need Google’s $250 per month AI Ultra plan to experience Project Genie. You will also need to live in the United States and be 18 years or older. At launch, Project Genie offers three different interaction modes: World Sketching, exploration and remixing. The first sees Google’s Nano Banana Pro model generate the source image that Genie 3 will use to create the world you’ll explore later. At this point, you can describe your character, set the camera perspective – whether first-person, third-person, or isometric – and how you want to explore the world that Genie 3 is about to generate. Before you can begin creating the model, Nano Banana Pro “sketches” what you are about to see so you can make changes. It’s also possible to write your own prompts for worlds that others have generated with Genie.

One thing to keep in mind is that Genie 3 is not a game engine. While its outputs may be game-like and simulate physical interactions, there are no traditional game mechanics here. Builds are also limited to 60 seconds, as is the presentation, capped at 24fps and 720p. Still, if you’re an AI Ultra subscriber, this is a great opportunity to see the cutting edge of what DeepMind has been working on over the past couple of years.