Spatial Insider
← NewsGenerative AI

Google Earth can now reimagine any place with AI.

Create image starts with the map on screen. AI turns it into a new scene that you can edit and save.

By Justin Ryan · · 2 min read

ShareXFacebookLinkedIn
The blue Google Earth globe changing into a satellite view of a city at night
Image: Spatial Insider with Google Earth

Bilawal Sidhu turned daytime Austin into a night scene without leaving Google Earth. In the same demo on Threads (opens in a new tab), he pictured the Colosseum in another era and saved the result to his map.

He did it with Create image (opens in a new tab), a new tool marked Experimental. It uses the place already on screen as the starting point for an AI picture.

Start with the map

Open a place and choose a view. Then type what you want to see. Google Earth keeps the location and camera angle while the AI creates a new image.

You can ask for another change, refine the result, or save it to a project. Sidhu called the feature Nano Banana. The interface uses a banana icon, but Google does not name the exact AI model.

The experimental Create image tool open over a 3D Google Earth view of Austin
Create image begins with the location and camera angle already on screen.Image: Spatial Insider via Google Earth

Picture what could be there

Create image makes a flat picture. It does not rebuild the live 3D map or add places you can move around.

Google warns that AI can make mistakes. An ancient version of the Colosseum may look real, but it is not a historical record.

The value is speed. A planner could show how a park might fill an empty lot. A designer could test a building at night. A storyteller could move a familiar place into another time.

Google Earth becomes a canvas

Create image is still labeled Experimental. Google has not said who will get it or when. Spatial Insider found it in the web app on a Google AI Pro account.

For now, it makes one picture at a time. That is still a big change for Google Earth. The map no longer has to show only what is there. It can also help people envision what could be.

More from Capture & Create.

See all stories
An Austin 3D map with colored rays connecting photos to their matched positions

Visual positioning · 2 min read

AI-powered visual positioning can find you from just one photo.

MultiSet matched an ordinary photo against a 3D scan of Austin. The visible scene led it back to the camera’s exact position.

An InSpatio demo showing source views beside a new view of a man reconstructed from one video

Spatial AI · 3 min read

InSpatio turns one video into a moving 4D Gaussian scene.

The research preview creates new camera angles from one video clip, without a multi-camera setup.

Adobe Premiere timeline and video editor arranged as floating panels in visionOS

Apple Vision Pro apps · 3 min read

Adobe Premiere now edits spatial video on Apple Vision Pro.

Premiere’s new visionOS app can combine flat and spatial clips, place titles in depth, and export a spatial video without leaving Vision Pro.