🛡️OpenAI Joins C2PA and Adds SynthID to Tag AI Images
TL;DR
OpenAI became a C2PA conforming generator and is adding Google's SynthID watermarks to images from ChatGPT, Codex, and its API. A public tool now lets anyone check whether an image came from OpenAI.
OpenAI became a C2PA conforming generator and is adding Google's SynthID watermarks to images from ChatGPT, Codex, and its API. A public tool now lets anyone check whether an image came from OpenAI.
Key Points
Now a C2PA Conforming Generator, so other platforms can read and preserve its provenance metadata
Adding Google DeepMind's invisible SynthID watermark, starting with images
A public verification tool in preview checks whether an image was made with OpenAI tools
Layered design: C2PA carries context while SynthID survives when metadata is stripped
Why It Matters
As AI images flood the web, cryptographic provenance paired with watermarks is becoming table stakes for platforms that want to label fakes.
Quick Facts
Frequently Asked Questions
Why does this matter?
As AI images flood the web, cryptographic provenance paired with watermarks is becoming table stakes for platforms that want to label fakes.
What happened?
OpenAI became a C2PA conforming generator and is adding Google's SynthID watermarks to images from ChatGPT, Codex, and its API. A public tool now lets anyone check whether an image came from OpenAI.
Comments
Be the first to comment
Enjoyed this article?
Get it daily. 7am. Free. Reads in 5 minutes.
Join 3,313 builders reading daily.