The Pitfalls of Integrating AI Image Generators into Google Earth

Unlocking Reality-Warping Images with Google Earth’s Nano Banana 2 Feature

A simple text prompt now has the power to create mind-bending images using Google Earth’s satellite, aerial, and 3D imagery. Renowned investigator Henk van Ess from Digital Digging showcases this technology with images depicting “refugees near the Mexican border” and a bomb crater near a hospital in Gaza.

In response to concerns over AI-altered images, Google assures users of its commitment to combating misinformation. Each image created with Nano Banana in Google Earth is embedded with a SynthID digital watermark for verification. Suspicious images can be cross-checked using the Gemini app or Lens in Search to determine their authenticity. Additionally, the use of Google’s “@verifyai” tag in Gemini, comparing images with other satellite platforms like Sentinel-2 or Landsat, and verifying orbital data are recommended by van Ess for validation.

Recently launched on the web version of Google Earth, the Nano Banana 2 feature is touted as a tool for visualizing historical sites and real estate projects. While some altered images may be easily identified as fake, there is a potential for misinformation to spread with believable creations. Despite Google’s watermarking efforts, Digital Digging managed to deceive Hive’s AI detector with an altered video from Google Earth.