Google has opened access to Project Genie, its experimental artificial intelligence (AI) project that helps create simple video games from pictures or descriptions. The prototype enables users to build their own interactive or 3D worlds.
Project Genie is a prototype web app powered by generative AI tools Genie 3, Nano Banana Pro, and Gemini. It can be used for storytelling and other creative projects.
Using a text prompt or images, users can create their own videos. They can design visual environments and develop characters that fit their narrative. To give users more control, Nano Banana Pro has been integrated into a feature called World Sketching, which allows creators to preview their work and make adjustments as needed.
The experimental project, which comes months after the preview of Genie 3, also enables creators to generate paths in real time and adjust camera angles based on their preferred point of view.
More advanced creators can remix existing worlds into new interpretations by building on their prompts. A gallery of curated worlds is also available, which users can explore for inspiration or reference.
Videos can be downloaded once creators are satisfied with their output.
Project Genie remains an experimental research prototype under Google Labs, with several limitations still in place. Genie 3 cannot yet recreate real-world locations with full accuracy, and modeling interactions among multiple independent agents within shared environments continues to be a technical challenge. Text elements are generated only when included in the input world description. The model also supports only a few minutes of continuous interaction rather than long sessions.
Project Genie is open to Google AI Ultra subscribers in the United States.
