Google has unveiled Project Genie, a cutting-edge AI-powered web app that lets users design, build, and explore their own interactive virtual worlds. The tool was introduced by Google CEO Sundar Pichai, who playfully described his own experience with it as “out of this world.”
What Is Project Genie?
Project Genie is a prototype web application powered by Google’s Genie 3 world generation model combined with Nano Banana Pro visuals and the Gemini AI platform. It enables users to create dynamic 3D environments based on text descriptions and uploaded images. The resulting worlds are not just static scenes — they can be explored from different perspectives and react to users’ movements in real time.
“World Sketching” — Start With Your Imagination
The creative process begins with what Google calls “World Sketching”. Users enter text prompts describing elements such as terrain, characters, or atmosphere — or even upload a photo of a physical object. Project Genie then interprets these inputs to generate a custom interactive world that users can immediately explore.
Exploring and Remixing Worlds
Once a world is generated, you can dive in from either a first-person or third-person perspective and interact with your environment. The AI continuously builds out the terrain as you move through it, simulating physics and dynamic reactions in response to your actions.
There’s also a remix feature that lets users take existing environments and modify them with new prompts, fostering creativity and collaboration among builders.
Limitations and Current Access
At present, Project Genie is available only to Google AI Ultra subscribers (18+) in the United States. World explorations are limited to 60-second sessions, and Google notes that the environments may not always render perfectly. Some experiences may feel less polished or have control latency.
