Google Launches Project Genie AI World Model for Creating Interactive Virtual Environments

Google releases Project Genie AI world model to U.S. subscribers, enabling users to create interactive virtual environments through text and images, explore real-time generated worlds, and remix existing creations, though current limitations include 60-second generation caps and control issues.

• Google launches Project Genie, an experimental AI world model prototype now available to Google AI Ultra subscribers in the U.S., allowing users to create and explore interactive virtual environments
• The platform is powered by Genie 3 technology and enables three core features: world sketching with text and images, real-time world exploration that generates paths as users move, and world remixing to modify existing creations
• The prototype currently has limitations including less realistic world generation, character control issues, 60-second generation limits, and higher latency, with plans to expand access to more territories over time