355
Audio & Video Production344
Automation & Workflow224
Software Development250
Marketing & Growth192
AI Infrastructure & MLOps173
Writing & Content Creation203
Data & Analytics140
Design & Creative169
Customer Support130
Photography & Imaging156
Sales & Outreach125
Voice & Speech135
Operations & Admin87
Education & Learning131
Google DeepMind is linking Street View with Project Genie so some users can simulate real streets, change weather, and test rare scenarios in a virtual space.
In short: Google DeepMind is connecting Google Maps Street View with Project Genie so people can simulate and explore real streets in an interactive virtual environment.
Google DeepMind announced a new integration between Street View and Project Genie at the Google I/O developer conference. Project Genie is a “world model,” which is an AI system that can generate a place you can move around in, like a video game world.
By adding Street View, Genie can build simulations anchored to real locations. Someone could explore a city block and also change conditions, like making it snowy or showing a more extreme scenario (like a flood), to see how things might look.
DeepMind researcher Jack Parker-Holder said this can help both people and machines. He gave an example of a robot in London that rarely sees bright sun. A simulation could include those rare sunny moments so the robot is not surprised by reflections and glare later.
Google said it has collected Street View data for about 20 years, with more than 280 billion images across 110 countries. The company is rolling out Street View in Genie to some Google AI Ultra subscribers in the US starting today, with wider access over time and global Ultra access over the next few weeks.
DeepMind also said the tool is still experimental. It is not fully accurate yet, and the simulations can look more like a video game than real life. It also does not reliably understand physics (cause and effect), such as a person bumping into objects.
Simulations like this can be a safer place to practice. That could help train robots and self-driving systems, and it could also give travelers and game makers a more interactive way to preview real places (like taking a “test walk” before you visit).
Source: TechCrunch AI