Via design.google
The Nano Banana 2 feature let users alter Google Earth imagery with text prompts, raising immediate questions about AI-generated misinformation and its implications for trust in digital media.
Google pulled the plug on an AI-powered image editing feature inside Google Earth roughly one day after launching it. The tool, called Nano Banana 2, let users modify satellite imagery using simple text prompts, effectively creating deepfakes of real-world locations with a few keystrokes.
Here’s the thing: it took less than 24 hours for someone to demonstrate exactly why that’s a problem.
What happened #
Nano Banana 2 launched on a Thursday as a new Google Earth feature designed to let users generate AI-altered versions of actual satellite images.
Digital ethics researcher Henk van Ess quickly put the tool through its paces. He used text prompts to generate images showing refugees near the Mexican border and a bomb crater next to a hospital in Gaza. Neither scenario was real. Both looked convincingly like authentic satellite imagery.
Google’s initial response pointed to two safeguards. Images generated with Nano Banana 2 included a digital watermark, and the company said it prevented “image creation on harmful topics.”
Earlier versions of the Nano Banana model were discussed in Google blog posts as early as February 2026, with the company touting significant improvements in speed, accuracy, and integration with real-time geospatial data. Nano Banana 2 was positioned as a substantial upgrade tied directly to satellite and aerial imagery sources. The ambition was clear: let users visualize locations historically or futuristically using AI grounded in real geographic data.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our