Google pulled its Google Earth AI tool one day after launch. The company said users shared screenshots that appeared to violate its policies.
Google introduced the “create image” feature globally on July 30. Nano Banana 2 transformed satellite, aerial and 3D imagery through text prompts.
The company announced the rollback on July 31 and promised stronger safeguards. It said generated outputs never appeared in the main Google Earth experience for other users.
Google said every output carried an AI watermark. Its SynthID technology lets users check images through Gemini or Google Lens.
However, AFP said its tests produced fake depictions of an explosion in Paris and a nuclear facility in Iran. The tool also generated a bomb crater in Russia and an Islamic State training ground in Syria.
Other prompts created fictional floods in Bangladesh and a migrant camp near the US-Mexico border, AFP reported.
Brady Africk analyses satellite imagery at the American Enterprise Institute. He warned that integration with real mapping data could make fabricated scenes more convincing.
Read: Google Photos Batch Editing Added by Chrome Extension
Digital investigations expert Henk van Ess said such images could inherit Google Earth’s credibility. Ross Burley of the Centre for Information Resilience warned that the feature could damage trust in satellite evidence.
Google’s generative AI policy prohibits misinformation, misrepresentation and misleading activities. The company did not announce when the feature might return.
