Google just dropped Nano Banana 2 into Google Earth, and honestly, it's the kind of feature that makes you want to spend way too long zooming around the globe.
Announced on July 30, the update brings Gemini's AI image generation straight to Google Earth on the web. The idea is simple: pick a spot, hit "create image," describe what you want to see, and Nano Banana 2 handles the rest. What makes this different from your typical AI image generator is that it's grounded in reality. Instead of starting from a blank canvas, the model uses Google Earth's satellite, aerial, and 3D imagery as its foundation. The results feel connected to the actual place, not just a random AI hallucination.
So what can you actually do with it? Google highlighted a few use cases that go beyond just messing around. Teachers can type something like "render a hyper-realistic view of what these Pompeii ruins looked like in 78 AD" and watch ancient history come to life in the classroom. Urban planners and architects can visualize projects before they break ground—imagine reimagining an empty Tokyo lot as a vibrant retail district. Or you can just have fun with it. One reviewer famously reimagined Philadelphia overrun with zombies and evil clowns.
It's available globally starting today for all Google Earth web users. That said, don't expect architectural blueprints. Early reviews suggest the tool is entertaining but not quite ready for serious professional work. And as with any AI, the outputs are interpretations, not factual documents.
Still, there's something genuinely exciting about a tool that lets you see what could be, not just what already is. Google Earth has always been about exploration. Now it's about imagination too.
Tags:
Technology