Google rolls out Project Genie to let users build AI generated worlds

Google has begun rolling out access to Project Genie, an experimental AI research prototype that allows users to create, explore and remix interactive digital worlds generated entirely by artificial intelligence.

The tool is currently accessible to Google AI Ultra subscribers aged 18 and above in the United States, with the aim to extend it to other regions in the future. Project Genie is hosted on the web by Google Labs and is driven by the Genie 3 world model, in addition to Gemini and Nano Banana Pro.

Google Project Genie

Project Genie gives users hands-on access to Google’s general-purpose world model, which is designed to generate dynamic environments that respond in real time to user actions. Unlike traditional AI image or video tools, Genie focuses on interactive spaces rather than static scenes.

Google said the prototype is built around three main features: world sketching, world exploration and world remixing.

With world sketching, users can generate environments using text prompts or images they upload. They can define landscapes, characters and even how movement works, such as walking, flying or driving. Nano Banana Pro allows users to preview and refine images before entering the world, while perspective options include first-person and third-person views.

World exploration enables users to move through these environments as they are created on the fly. As users navigate, Project Genie predicts and generates what lies ahead based on their actions, allowing for continuous movement through the AI-generated world. Camera angles can also be adjusted during exploration.

The third feature, world remixing, lets users modify existing worlds by changing the underlying prompts. Google has also added a gallery of curated worlds and a randomiser tool to help users find inspiration. Finished creations can be exported as video clips.