Google launched Project Genie on January 29, 2026, for Google AI Ultra subscribers in the United States who are 18 years or older. This experimental research prototype, housed in Google Labs, allows users to create, explore, and remix interactive worlds using text prompts and images. The web app is powered by the Genie 3 world model, Nano Banana Pro, and Gemini.
Project Genie builds on Genie 3, a general-purpose world model previewed in August 2025. Genie 3 generates diverse, interactive environments. Trusted testers previously created various worlds with it. The prototype provides hands-on access to this technology, focusing on immersive world creation.
World models simulate environment dynamics, predicting evolution and the impact of actions. Google DeepMind has developed agents for games like Chess and Go. Project Genie advances toward systems that handle real-world diversity for general AI.
The prototype features three core capabilities. World sketching enables users to prompt with text and generated or uploaded images to build a living, expanding environment. Users create a character and define exploration modes, such as walking, riding, flying, or driving. Integration with Nano Banana Pro allows preview and fine-tuning. Users select first-person or third-person perspectives.
World exploration turns the environment navigable. As users move, Project Genie generates paths ahead in real time based on actions. Camera views adjust during traversal.
World remixing lets users build on existing worlds. They access a gallery of curated worlds or use a randomizer for inspiration, then modify prompts. Users download videos of their worlds and explorations.
Diego Rivas, Product Manager at Google DeepMind; Elliot Breece, Product Manager at Google Labs; and Suzanne Chambers, Director at Google Creative Lab, contribute to the project.
Genie 3 powers the prototype as an early research model. Known limitations include worlds that may not fully match prompts, images, or real-world physics, lacking complete realism. Characters sometimes resist control or show higher latency. Generations cap at 60 seconds.
Some Genie 3 features from the August announcement, such as promptable events that alter the world during exploration, remain unavailable in this prototype. Details on limitations and updates appear on dedicated Google pages.
Access rolls out to U.S. Google AI Ultra subscribers aged 18 and above, with expansion to other territories planned. Google seeks feedback from these advanced AI users to understand applications in AI research and generative media. The company intends to widen access to Project Genie and its world-building technology over time.
Project Genie operates responsibly as part of efforts toward general AI systems that benefit humanity. It stems from work with trusted testers, now extended to subscribers.
