Google Earth’s AI deepfake tool only lasted one day

Google has withdrawn a newly launched Google Earth feature that enabled users to edit satellite images using text prompts powered by artificial intelligence. The tool allowed the creation of realistic deepfakes depicting sensitive scenarios, such as refugees at borders or bomb craters near hospitals, raising significant concerns about misinformation. Although Google initially defended the feature by highlighting digital watermarks and safety filters, security researchers successfully bypassed detection tools and generated misleading content without restrictions. Consequently, Google decided to roll back the feature to develop stronger guardrails, acknowledging the high level of trust users place in Google Earth for reliable geographical information.

(Source：The Verge)