Google Earth's New AI Lets Anyone Fabricate Misleading Satellite Imagery
404 Media · rss · 2026-08-01
Google has integrated a new AI feature into Google Earth, powered by its Nano Banana model, allowing users to generate fabricated buildings or scenarios directly on the satellite map via text prompts.
Tests revealed that users could easily fake highly misleading images, such as drone strike ruins, refugee camps, or nuclear plants. While Google claims to have implemented SynthID watermarks and blocks on harmful topics, OSINT researchers warn this will severely undermine the credibility of satellite imagery and facilitate the spread of disinformation.
More from Multimodal
- ByteDance Releases Seedance 2.5: Native 30-Second AI Video Generation — hey_abusiddik · 2026-08-01
- Seedance 2.5 Video Generation Demo Shows Next-Level Fabric Physics — future_coded · 2026-08-01
- Advanced AI Video Prompts: Direct Like a Shot List, Not a Wish — HeyAmit_ · 2026-08-01
- Seedance 2.0 Price Drop on Dreamina: $0.083 Per Second — FellMentKE · 2026-08-01
- AI Video Workflow Breakdown: Crafting Consistent Multi-Character Shorts — LudovicCreator · 2026-08-01
- Seeking Local ComfyUI Workflows for Google Earth-style Satellite Zoom-ins — justlooking___1 · 2026-08-01