Google shuts down Google Earth's AI image editing tool after one day
Google has pulled the plug on a new Google Earth feature that allowed users to alter satellite images with text prompts, citing policy violations and a need for stronger safeguards.
Google on Thursday launched a feature in Google Earth that let users edit satellite images using AI text prompts, but by Friday it was gone. The tool effectively enabled users to create AI deepfakes of the real world by modifying imagery based on written requests.
Journalist Henk van Ess used the feature to deliberately generate images showing refugees near the U.S.-Mexico border and a bomb crater next to a hospital in Gaza. Google's initial response on Thursday said that images produced with Nano Banana 2 in Google Earth included a digital watermark and that the company prevented "image creation on harmful topics."
A day later, Google reversed course. In a new statement, the company acknowledged that people trust Google Earth for an accurate view of the world. While geospatial professionals had used the feature for useful purposes, Google said it saw users sharing screenshots of generated imagery that appeared to violate its policies. As a result, the feature is being rolled back while stronger guardrails are implemented.
Google also noted that the generated images did not appear in the main Google Earth experience for others to see, and they were watermarked as AI-generated. However, van Ess demonstrated that a video made with Google Earth fooled Hive's AI detection tool, showing how such media could spread despite the watermarks. He said that during his testing, "Nothing was refused, nothing was softened, and nothing suggested I try a different prompt."
