Top AI Product

Google shipped a feature, then killed it in under 24 hours. On Google Earth for web, you could zoom to any spot, tap “create image,” type a prompt, and get an AI-generated satellite view of whatever you imagined. By the next day it was gone.

What it was, and why it broke

This wasn’t a chatbot or a coding tool — it was one-tap generative imagery bolted onto the world’s most-trusted map. That’s the problem. Satellite imagery is the last thing OSINT researchers trust to verify atrocities in places reporters can’t reach. Open-source investigator Henk van Ess, who first sounded the alarm, prompted refugees at the Mexican border, a nuclear plant in Iran, a bomb crater on a hospital in Gaza. Nothing was refused. NPR generated Iran’s Kharg Island on fire just as easily.

The real story

Google admitted it saw users sharing screenshots that violated its policies, then rolled back. Classic ship-first, patch-later AI safety miss — except here the thing being faked is ground truth itself.

You Might Also Like

• Google Releases Gemini Embedding 2 one Vector Space for Text Images Video and Audio
• Google Deepmind Aletheia Just Solved Math Problems Nobody Could Heres why That Matters
• Google Workspace Studio Just Made ai Agents a Thing Everyone can Build
• Google A2ui Agent to User Interface Finally a Standard way for ai Agents to Show you Things
• Google Lyria 3 Just Turned Gemini Into a Music Studio and im Weirdly Into it