Google Abruptly Withdraws Google Earth AI Image Tool After Disinformation Backlash

Google has pulled a newly introduced generative AI feature from Google Earth just 24 hours after its launch following intense backlash from open-source intelligence researchers and journalists who warned it enabled the rapid creation of fabricated satellite imagery. Powered by Google’s Nano Banana 2 image-generation model, the tool allowed users to select real-world geographic coordinates and apply text prompts to superimpose photorealistic elements—such as fake explosions, military strikes, disaster zones, and structural damage—directly onto the platform’s aerial and 3D map data. While Google originally pitched the functionality for creative urban planning, historical visualizations, and architectural design, investigators quickly demonstrated that the system lacked effective prompt restrictions, allowing users to effortlessly generate and screenshot convincing fake scenes. In a statement acknowledging the reversal, Google noted that while generated images carried digital watermarks and were not visible on the main public map to other users, screenshots shared on social media violated platform policies and threatened to undermine public trust. The tech giant confirmed it is pausing the feature while working to implement stricter safety guardrails, emphasizing that Google Earth remains a critical global reference point for verifiable geospatial truth.