Google has launched Project Genie, a tool that allows users to create three-dimensional virtual environments through simple text prompts. Currently available in the U.S. for Google AI Ultra subscribers at $250 per month, the service offers enhanced features like 30 terabytes of cloud storage and higher AI usage limits. The underlying model, Genie 3, can render interactive environments at a resolution of 1280 by 720 pixels and up to 24 frames per second. Users interact with Project Genie by entering descriptions for both the 3D environment and the avatar navigating it. The tool utilizes Nano Banana Pro, an image generation model that can transform basic sketches into photorealistic 3D objects. Once a virtual world is generated, users can further customize it or choose from pre-packaged designs. Interactions are limited to 60 seconds, though the technology has the potential for longer sessions in future updates. Users can also save their virtual experiences as videos. Google plans to expand Project Genie to other markets and may eventually offer it through an API for developers. The rendered environments have applications in generating visual training data for AI projects, enhancing their utility beyond just entertainment or exploration.