MOUNTAIN VIEW: Google has taken down a newly launched AI feature in Google Earth after users pushed its limits by creating realistic-looking fake scenes of wars and disasters, sparking concerns over the spread of misinformation.
The feature, powered by Google’s Nano Banana 2 AI model, allowed users to generate photorealistic images through text prompts using Google Earth’s satellite, aerial, and 3D imagery.
Google introduced the tool for professional applications, including education, urban planning, and design visualisation. However, shortly after its launch, users demonstrated that it could also be used to create fabricated scenes resembling real-world events, including conflicts, bomb damage, floods, and refugee camps.
Google said it had observed professionals using the tool productively but also identified cases where generated images appeared to violate its policies.
The company clarified that AI-generated images created through the feature carried watermarks and were not displayed publicly within the Google Earth experience.
Read More: EU imposes major fine on Google for anti-competitive conduct
However, Google acknowledged that screenshots of fabricated images could still circulate online and potentially contribute to misinformation.
Following the concerns, Google temporarily removed the feature and said it is working on additional safeguards to ensure the tool is used responsibly and within legal guidelines.
The company has not announced a timeline for bringing the feature back but said improvements are being developed before a possible relaunch.
The incident highlights the growing challenge for technology companies in managing advanced AI tools that can create highly realistic content and the risks associated with their misuse.
