Google has rolled back a newly launched AI image-generation feature in Google Earth just one day after introducing it, following the circulation of user-created images that appeared to breach company policies. The feature, powered by Google's Nano Banana 2 model, let users generate photorealistic images from Google Earth's satellite, aerial and 3D imagery using text instructions, for locations across the world.

The tool quickly drew criticism over its potential to spread misinformation: by using real-world landmarks and monuments as the base imagery, it made it possible to fabricate plausible scenes at recognizable locations. In a post on X, Google said it had seen geospatial professionals using the feature for a range of practical applications, but that users were also sharing screenshots of generated imagery that appeared to violate its policies.

"We know that people uniquely trust Google Earth for a reliable view of the world," the company said. "So we're rolling back this feature in Google Earth while we work on implementing stronger guardrails." Google stressed that generated images were watermarked as AI-created and were not displayed within the main Google Earth experience for others to see.

The suspension is the latest example of tech companies scaling back AI features over misuse concerns. Earlier this month, Meta discontinued Muse Image, an AI tool that generated images from public Instagram accounts, after criticism over privacy — including objections from a Hollywood union.

The episode highlights a growing tension for platform makers: generative-AI features are powerful differentiators, but when they touch trusted, factual products like maps, the cost of a policy misstep is measured in user trust.

Sources