Google launches Project Genie allowing users to create interactive AI-generated worlds
The experimental research prototype rolls out to Google AI Ultra subscribers in the United States, with plans to expand access to additional territories.
Google has launched Project Genie, an experimental prototype that allows users to create and explore interactive AI-generated worlds. The web application, powered by Genie 3, Nano Banana Pro, and Gemini, is rolling out to Google AI Ultra subscribers in the US aged 18 and over.
Genie 3 represents a world model that simulates environmental dynamics and predicts how actions affect them in real time. Unlike static 3D snapshots, the technology generates paths in real time as users move and interact, simulating physics for dynamic environments.
Project Genie centres on three core capabilities: world sketching, exploration, and remixing. Users can prompt with text and images to create environments, define character perspectives, and preview worlds before entering.
As users navigate, the system generates paths in real time based on their actions.
The experimental prototype has known limitations, including generation restrictions to 60 seconds, potential deviations from prompts or real-world physics, and occasional character controllability issues.
Google emphasises responsible development as part of its mission to build AI that benefits humanity, with ongoing improvements planned based on user feedback.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!