Google Pulls Earth AI Image Tool After Deepfake Concerns Spark Backlash

Google has withdrawn its newly launched AI image generation feature for Google Earth less than 24 hours after making it available to users, following widespread concerns about its potential to create convincing fake satellite imagery. Powered by Google’s Nano Banana 2 AI model, the feature allowed users to generate realistic images of real-world locations using simple text prompts and overlay them onto satellite and 3D map views. While Google introduced the tool for educational, creative, and visualization purposes, it was quickly used to create fabricated images of military sites, disaster zones, and other sensitive locations, prompting criticism from researchers and the open-source intelligence (OSINT) community.

The feature was designed to help users visualize ideas by combining AI-generated content with Google Earth’s mapping platform. However, critics warned that integrating AI-generated imagery into a service widely trusted for geographic accuracy could blur the line between authentic satellite imagery and fabricated content. Within hours of its release, users shared examples of realistic-looking but entirely fictional scenes, including bomb craters, nuclear facilities, and humanitarian crises, raising fears that the technology could be exploited to spread misinformation or manipulate public opinion.

In response to the backlash, Google confirmed that it was “rolling back” the feature while it develops stronger safety measures. The company noted that AI-generated images included its SynthID digital watermark and were not displayed as part of the core Google Earth imagery. Nevertheless, experts argued that watermarks alone were insufficient because screenshots and edited versions could still circulate online without clear indications that they were AI-generated. Google said it plans to improve its guardrails before considering a future relaunch of the feature.

The controversy has reignited debate over the responsible deployment of generative AI technologies. Researchers, journalists, and fact-checkers warned that realistic AI-generated satellite imagery could undermine the work of investigators who rely on geospatial evidence to verify conflicts, natural disasters, and environmental changes. Many experts believe mapping platforms require stricter safeguards than conventional image generators because they are often treated as trusted sources of factual information.

Google’s swift decision to suspend the feature highlights the growing challenges technology companies face as AI capabilities become more powerful. While generative AI offers new creative possibilities, it also increases the risk of misinformation when integrated into trusted platforms. The incident is likely to influence how AI image generation tools are developed in the future, with greater emphasis on transparency, content authentication, and safety measures before large-scale public releases.