Google DeepMind has integrated data from Google Street View into its Project Genie to simulate diverse, interactive real-world environments. The new feature debuted at the Google I/O developer conference in May 2026 and is now available globally for adult Google AI Ultra subscribers [1, 2].
Project Genie allows users to create interactive 3D environments based on actual locations. Users can control cameras and avatars within these environments, which can last up to 60 seconds and stream at 720p resolution at 24 frames per second [2].
The integration leverages over 280 billion Street View images collected over two decades, covering 110 countries across all seven continents [1]. This vast dataset enables Project Genie to simulate real-world conditions such as weather variations, supporting use cases like robotics training and educational experiences [1].
Project Genie 3 launched as a research preview in August 2025 and later opened to United States Google AI Ultra subscribers in January 2026 before expanding globally this May [1, 2].
Jack Parker-Holder of Google DeepMind said, "It’s really powerful for both the agent [and robotics] use case and for humans to play with, and that’s always been the thesis of Genie" [1].
There is some debate over Project Genie's purpose. While Google states it supports gaming, education, and robotics by creating interactive worlds from text or images [1], other sources suggest it is not designed as a full game generator and lacks features like consistent mechanics and narrative flow required for true game development [2].
Users can now access Project Genie’s new Street View feature worldwide through their AI Ultra subscription, adding real-world context to virtual simulations for various applications [1, 2].
