What you need to know
- Google’s Project Genie development continues, and at I/O, the company unveiled its newest merge with Google Maps’ Street View.
- Users can click on maps, select a place in the U.S., and reimagine that place in a new way with a prompt (think Chicago as a desert).
- Google says Street View is rolling out gradually to AI Ultra subscribers today (May 19) for places in the U.S.
This project name might sound familiar if you tuned in last year to I/O; regardless, Google is announcing the next steps for Project Genie “in the real-world.”
Simulations are a pretty straightforward concept, and Google’s continuing its Project Genie ambitions by merging your imagination with what’s real. The company said during I/O that it’s “taking a leap forward by connecting Genie’s generative power with real-world imagery of Google Street View…” Project Genie is still the company’s prototype; however, it’s infusing Street View’s “grounding capabilities” into the AI model to put a “creative twist” on what’s in our world.
What users will soon find is the ability to create an imaginary world in Genie with the backbone of a real place as its base.
Street View is a recognizable name, as it’s one of the primary features in Google Maps. With this Genie update, users can “tap the Maps pin to choose a place in the U.S. and optionally select a style for your world.” This last bit is the imaginary aspect. If you’re clicking on Chicago, you can try turning it into a barren wasteland, marred by time and sandstorms. Moreover, you can add a character to try to bring the scene together.
Beyond just environmental changes, Google teases that users can turn a city black and white with vintage cars and more. Maps Imagery Grounding carries this feature forward. It’s software that developers reportedly utilize to blend an AI-generated creation with Street View.
Genie sprinkles more AI
Genie’s Street View is available today (May 19) for places in the U.S. Google states it plans to expand this in the future. With that, Google AI Ultra users will see this feature gradually arrive starting today.
Project Genie made a brief appearance during I/O 2025. Genie 3, the company’s AI model in the spotlight, was highlighted as a feature that would let users create a 3D world based on a written prompt. That wasn’t all, as Google also stated users could then “traverse” or “explore” this generated world. What you’re dropped into is a 720p, 24FPS world that the AI generated based on your description.
This simulated world would respond to your actions and would even survive for over a minute before giving out. Similar to how Google views Genie today, this is still an experiment, and things could get a little wonky. Still, the company is interested in advancing its AI simulation tech.
Android Central’s Take
This gives a little bit of intrigue. To be able to “see” what a real city would look like in a totally different light. Like Chicago as a wasteland that makes you think of Mad Max. Maybe London is posing as Atlantis. Of course, it’s just an AI-generated creation, so it’ll lack some of that finesse, but it might be a cool thing to mess around with when there’s a friend or a family member around. Just to get crazy and see what you can mess with.