SynthID Detector(synthid.com) |
SynthID Detector(synthid.com) |
Even Google themselves previously offered a no-auth way (just very niche): if you uploaded an image to Google Image search, and went to "About this image", it would show if the image was Google AI-generated. Now it doesn't.
https://brand.io/article/spymarks/
We need to inform everyone that this technology can encode database identifiers and enough entropy to uniquely identify you as an author (or downloader).
This extends to other forms of media as well.
It's not really possible to ban steganography, is it?
> That future lies halfway between now and 1984. So let’s stay off that timeline, shall we?
> Call a spymark what it is. A spy tool used to spy on you and everyone you interact with.
How hard can it be to download a set of real images and generated ones in order to train a model to detect and strip the watermark with minimal perceptual difference?
Given that it's an ad company, they want something to connect what you're doing to the rest of their data, and they couldn't possibly keep the old image search results there if they want people to use this.
The textbox below it says Paste image link, but you can actually paste an image from your clipboard here, too.
URLs expand Googlebot’s indexes; pasted images don’t.
This is the best SynthID write-up I've found so far: https://fyx.me/articles/attempting-model-extraction-of-googl...
It covers how it actually works (probably), and how to train your own classifier for it, with some seemingly decent results.
"Excellent work acquiring that outlook data Anat, now let's cross check the defunct company emails against YouTube videos edited with Google PrivateEditAI™ to figure out what the social security number of this YouTube account is."
> classifying images in bulk or offline
You've described exactly the elements spammers and fraudsters need to be able to defeat this mechanism.
(While a sibling comment points out putting the image through Google Image Search as an alternative, I don't remember this being signposted in the support article I've read, so I unfortunately didn't know about it)
https://help.openai.com/en/articles/8912793-provenance-signa...
There is rate limit though
In many situations, photos are evidence. AI tools make convincing fakes, such as images of defect product.
>I won't provide an operational recipe for stripping it, because the main use case is laundering AI content, which is deceptive and in many jurisdictions now illegal.
More detail at https://deepmind.google/models/synthid/
For those who, like me, were hoping it was a vision model to identify synthesizer models from photos of concerts and music studios!
Or did it just become public
Edit:
Blog post today https://blog.google/innovation-and-ai/models-and-research/go...
Hey Gemini, find a synonym for every second adjective. Replace in text.
Hey Grok, find a synonym for every third proper noun. Replace in text.
Hey …
Too short for watermarking but point taken of course.
Probably more like
response = frontier_synthid_model(prompt)
while frontier_synthid_validator(response) == false
response = cheapo_chinese_model.prompt(
"pls remove synthid",
response
)https://deepwalker.xyz/blog/evaluating-synthid-watermark-rob...
I suppose it was too convenient.