Google has withdrawn a newly launched artificial intelligence feature from its Google Earth platform less than 24 hours after it became available to the public. The tool, which utilized Google’s Nano Banana 2 image-generation technology, allowed users to zoom into any location on the map and use text prompts to create photorealistic images layered over real satellite, aerial, and 3D-mapping data. While Google initially promoted the feature as a way for users to visualize history or plan real estate projects, it was quickly removed following widespread criticism from researchers and open-source intelligence experts.
Critics demonstrated that the tool could be easily manipulated to generate convincing but entirely fabricated scenes. By inputting specific prompts, users were able to create images of non-existent events or structures, such as nuclear power plants in Iran, refugee camps at the U.S.-Mexico border, and damaged infrastructure in conflict zones. These synthetic images could be anchored to real-world coordinates, creating a high risk for the spread of misinformation. Experts warned that because Google Earth is widely considered a reliable source of geographical truth, these AI-generated fakes could be used to mislead the public or fuel geopolitical tensions.
In response to the backlash, Google issued a statement acknowledging that while many professionals found the tool useful, some users were sharing screenshots that violated the company's policies. The company emphasized that the generated images were watermarked as AI-created and did not appear in the main, public-facing Google Earth experience. However, the ease with which these images could be captured and shared on other social media platforms prompted the immediate rollback. Google has not provided a timeline for when the feature might return or what specific safety guardrails will be implemented to prevent future misuse.