Google withdraws its new Earth AI image tool within 48 hours after testers generate convincing fake satellite imagery
Category: AI Governance • Trust & Safety
What it is
Google launched an AI image-generation feature in Google Earth on July 30, letting users overlay photorealistic, prompt-based scenes onto real satellite maps using its Nano Banana 2 model. Within 48 hours, testers including BBC Verify generated convincing fabricated imagery – a collapsed Eiffel Tower, tanks in Kyiv, a fake nuclear facility in Iran. Google pulled the feature, citing policy violations, and said it would return only with stronger safeguards in place.
Why it Matters for Enterprises
Trusted data sources become a misinformation attack surface the moment generative tools are layered on top. Enterprises embedding third-party AI features should stress-test for misuse before launch, not after.
