12
Identifying (google)AI-generated images with SynthID - [Your Thoughts?]
(deepmind.google)
Synth ID operates by embedding a digital watermark into AI-generated images, which serves as a hidden marker. This watermark is imperceptible to the human eye but can be detected using specialized tools. The tool is currently available exclusively to users of Google's AI image creator, Imogen.
https://youtu.be/9btDaOcfIMY?si=vK4bqDS3JOu5bl9H
https://deepmind.google/technologies/synthid/
https://www.theverge.com/2023/8/29/23849107/synthid-google-deepmind-ai-image-detector
With SynthID, you should be able to edit a photo all you want without destroying the AI watermark. Image: Google
It remains a dangerous dead end. Any competent fraud will remove the watermark, or use a generator that doesn't add one. Giving people the idea that the absence of a watermark makes something trustworthy, can only help bad actors.
I can agree on that. Someone can also add a watermark to a real photo to cause disinformation.