Google DeepMind connects Street View to Project Genie world model | TNW

Project Genie is connected to Google Street View imagery to create interactive, AI-generated simulations of real places. The system can use more than 280 billion images from 110 countries across all seven continents. Users can enter simulated environments such as snow-covered New York City blocks or London streets under rare sunshine and navigate them in real time. The feature is rolling out to Google AI Ultra subscribers in the United States, with broader global expansion planned. Genie 3 also supports robotics use cases, including powering a Waymo simulator used to train self-driving cars on rare scenarios.

"Project Genie, the company's general-purpose system for creating interactive environments, can now draw on more than 280 billion images captured across 110 countries and all seven continents. The result is a tool that lets you drop into a simulated version of, say, a New York City block covered in snow, or a London street bathed in rare sunshine, and navigate it in real time."

"Google DeepMind has connected its Project Genie world model to 20 years of Street View imagery, allowing users to wander through AI-generated simulations of real places. The integration, announced at the Google I/O developer conference on Monday, marks one of the most tangible demonstrations yet of what generative world models can do when paired with a colossal real-world dataset."

"Genie 3, the latest iteration of the model, first appeared as a research preview in August 2025, part of a broader push by Google to embed AI across its platform stack. In January 2026, DeepMind opened access to Google AI Ultra subscribers in the United States. The Street View integration is now rolling out to some Ultra users in the US, with a global expansion planned over the coming weeks."

"Jack Parker-Holder, a research scientist on DeepMind's open-endedness team, framed the feature as serving two distinct audiences. On one hand, robotics developers could use it to train agents in simulated environments that mirror actual locations. On the other, ordinary users could simply explore for fun."

Read at TNW | Artificial-Intelligence