AI Image Metadata: What ChatGPT and Gemini Leave in a File
We downloaded images from ChatGPT and Gemini and read their metadata: a signed C2PA manifest and a declared invisible watermark. What each one contains.
AI detection & provenance
AI image detection: reading metadata fingerprints
AI generators leave fingerprints in image metadata: C2PA manifests, Stable Diffusion parameters, ComfyUI workflows. Here's how to find them, and what their absence does (and doesn't) prove.
AI & provenance Can You Remove the Gemini Watermark? The Three Marks, Explained
Gemini images carry three marks: a visible sparkle, a C2PA record and an invisible SynthID watermark. Which ones you control, and which ones stay.
AI & provenance What Is SynthID? Google's AI Watermark, Explained
SynthID is the invisible watermark Google adds to content made with its AI. How it works, how to check an image for it, and why metadata tools cannot see it.
AI & provenance What AI image detectors can and cannot prove
Three kinds of evidence say an image was generated, and they are not equally strong. How to read a detector score, and the mistake almost everyone makes.
AI & provenance What is C2PA? Content Credentials explained
C2PA attaches a signed record of where an image came from. What that record proves, where it sits in the file, and the three things it cannot tell you.