Top
Best
New

Posted by smooke 8 hours ago

OpenAI Adopts Google's SynthID Watermark for AI Images with Verification Tool(openai.com)
224 points | 123 comments
827a 22 minutes ago|
Interesting that it seems to be the case that SynthID has been totally busted open, but OpenAI's new watermark has not yet [1]

[1] https://github.com/wiltodelta/remove-ai-watermarks

himata4113 4 hours ago||
if you tell it to generate the AI image with a black background you can visually see the synthid with a good enough monitor, it's just a repeating fuzzy pattern, nothing special.

I have found great success of getting rid of it by masking every 2nd pixel, regenerating missing pixels and then once again masking every 2nd pixel offset by 1.

Used an off the shelf model to fill in the pixels, but I also exported a depthmap first (before any alternations) and denoised it so generated masked pixels comform to the original content. The result was obviously not 100% perfect, but with more time and a model fine tuned for this specific use-case would be able to remove any kind of ai watermarking without too many issues.

teravor 4 hours ago||
i wouldn't have any confidence in being able to remove a 0.5 bit watermark (presence/absence). what you see is probably a functional decoy.
huflungdung 3 hours ago||
[dead]
cryptoegorophy 31 minutes ago|||
Can an image just be stretched or compressed a very tiny bit?
userbinator 3 hours ago|||
but with more time and a model fine tuned for this specific use-case would be able to remove any kind of ai watermarking without too many issues.

Always amusing to see AI used against itself.

m00dy 1 hour ago|||
It’s definitely hackable, Some of our engineers worked on this long time ago

https://deepwalker.xyz/blog/bypassing-synthid-in-gemini-phot...

tantalor 4 hours ago|||
But why
noir_lord 4 hours ago||
Because we are on Hacker News.
tantalor 4 hours ago||
good point
sigbeta 1 hour ago||
[dead]
big_toast 7 hours ago||
What information is included in the metadata or SynthID? How many bits can be encoded in a SynthID?

Can it be used to create something like nutritional labels for synthetic content? 10% synthetic text, 30 synthetic images.

Your reality was 15% synthetic today (75% mega corp, 25% open-weight neocloud).

big_toast 6 hours ago||
I guess the SynthID-Image paper from Oct 2025[0] was an encoder-decoder for which they tested checking a flag or a 136 bit payload in 512x512 images and the watermark's robustness after various transformations.

Presumably the deployed version is meaningfully different.

[0]:https://arxiv.org/html/2510.09263v1

echelon 5 hours ago||
This is very similar to audiowmark

https://github.com/swesterfeld/audiowmark

You can stuff per-item database unique IDs, user IDs, geohashes, and other nefarious things inside.

We need to protest this LOUDLY.

Our devices are being locked down, we're having attestation and trusted computing forced on us, the internet all over the world is undergoing age verification with full ID verification.

Just because this is on "ai images" today doesn't mean it won't be on all images - screenshots, your camera reel, etc. - in the fullness of time.

This is scary.

These are the tools of 1984. They've been boiling the water slowly, but in the last year things have really started to pick up pace. Please push back. Loudly.

Everyone at Google and OpenAI working on this: WHAT THE FUCK ARE YOU DOING. STOP.

We have laws and mechanisms to prevent revenge porn, CSAM, defamation, etc. They are robust and can be made even stronger. We do not need to sacrifice the security of our privacy and our speech to fight imagined harms when the real danger is turning into an authoritarian society.

tadfisher 3 hours ago|||
The point of SynthID is to make generated images identifiable, in an attempt to prevent 1984-esque situations where you can't believe your eyes and ears. Applying it to screenshots and camera output defeats its only purpose.

If the powers-that-be want to enforce age verification, watermarking camera output is not the correct technology to do so. It would be something like HDCP, where camera manufacturers are given keys and a whole trusted media path is built so that the relying party can cryptographically enforce that a trusted camera is being used to capture live images.

raron 2 hours ago|||
> The point of SynthID is to make generated images identifiable, in an attempt to prevent 1984-esque situations where you can't believe your eyes and ears.

You can still use traditional methods to manipulate images, too, so I don't think a "does not contain SynthID watermark" means you can trust that image more. In the other hand, encoding a lot of personal and other information in the watermark (136 bit is a lot) that can not be easily removed and most of the people are unaware of it seems really an 1984-like dystopia.

fc417fc802 1 hour ago|||
You have missed the point by such a wide margin that I have to wonder if it wasn't intentional.

The same techniques used here can be applied in other domains for other purposes. That would not "defeat its only purpose". The danger is the normalization of watermarking for [ insert good reason here ] with regulation eventually making it mandatory once everyone is accustomed to it. Rinse and repeat to gradually boil the frog.

We live in a world where nearly all printers already watermark everything they print with their serial number. It wouldn't be at all surprising if the next modernized variant of that technology encoded personal and contextual data tied to the user.

Dylan16807 1 hour ago||||
I'm going to save my protests for anyone trying to watermark real images.

Zero watermarks is a lot worse than semi-effective AI watermarks.

fc417fc802 1 hour ago||
Is it? Given local models this delays the current cutting edge at any given time by what, 6 to 12 months at best?
Dylan16807 1 hour ago||
Well the person I replied to seems to think it'll be at least semi-effective.
fc417fc802 1 hour ago||
How effective something is as an authoritarian tool (or whatever arbitrary purpose) can be (and very often is) completely unrelated to its effectiveness for some other unrelated purpose. It isn't clear to me why even highly effective AI image watermarks would be better than zero watermarks given what I pointed out about local models.
Extropy_ 5 hours ago||||
Most cameras already produce metadata. You can remove this metadata. Can you not also detect and remove watermarks?
big_toast 5 hours ago|||
The paper references some threat models they considered. They suggest someone might "possess paired information (both original and watermarked content)" and therefore be able to undo watermarking. Presumably it's fairly easy to get identity operations out of image APIs that would result in this situation. I'm not sure that addresses echelon's main concerns though.
alterom 4 hours ago|||
The metadata is kept separately from the original data, and is, by design, modifiable and removable.

Watermark, by design, irreversibly modifies the original data, and is, by design, hard to remove without producing detectable artifacts (or rendering the data useless altogether).

In short, the answer is no.

gumby271 3 hours ago|||
Or perhaps a user id or fingerprint to an individual. We added that to printers long ago, this would easily enable that for every photo and image you generate too.
janalsncm 5 hours ago||
Don’t think that would be possible. If I paste a synthetic piece into an otherwise organic image, the synth id isn’t going to know that.
animal_spirits 5 hours ago||
Synth ID can detect parts of images with the watermark.
WhatIsDukkha 6 hours ago||
This is just performative nonsense.

As someone that creates things with tools with different media I would just hard avoid this tool that adds...

arbitrary metadata not of my choosing.

Should I seriously make a texture for a videogame with this weird DRM glorp in it?

How old is photoshop and why is it exempt?

ericpruitt 6 hours ago||
Just because something isn't perfect doesn't mean it's not useful. I've already seen posts online that were able to be proven as falsified because someone ran the images through Google for SynthID checks.

> How old is photoshop and why is it exempt?

For one, it's not developed by Google or OpenAI. The barrier to entry to making realistic but deceptive images with Photoshop is far higher than with AI, and there are already techniques that can, imperfectly, be used to detect the use of traditional image editing.

WhatIsDukkha 4 hours ago|||
So 999 people that are just making an image need to be DRM'ed so that you might catch the 1 person making "realistic but deceptive" images... like this is some kind of special case of ... internet images.
space_fountain 4 hours ago||
This isn't DRM right? This is metadata attached to the image that makes it clear it was synthetically generated. The public has a huge incentive to know when images are AI generated and the harm to legitimate users seems pretty small: aka someone might complain online that you use AI
raron 2 hours ago|||
Not yet, but it is easy to imagine many ways it would be used for DRM.
WhatIsDukkha 4 hours ago|||
billions? of "fake" images not generated by ai but just photoshopped and ... not really harmful.

There is no case that any of its particularly harmful outside of things like CSAM which is illegal.

nathancahill 2 hours ago||||
I mean I see a lot of images online where people forget or don't care enough to remove/crop the Gemini watermark.
bsder 3 hours ago|||
I guarantee this works poorly, at best.

If this actually works solidly, Google is in deep, deep, deep shit. It would mean that I can put a mark on my non-AI videos and demand that Google not allow upload of my identifiably copyrighted content.

This would completely obliterate YouTube.

fc417fc802 1 hour ago||
No, it wouldn't. ContentID is already used by Google for that exact purpose. They appear to be fully in favor of enforcing IP law provided the owning party raises a complaint.
Jtarii 5 hours ago|||
>How old is photoshop and why is it exempt?

I'm sure you can think of a couple things that differentiate gen AI from photoshop, I believe in you.

WhatIsDukkha 4 hours ago||
The main difference is we are in the middle of a moral panic and people have lost perspective.

Its a tool with different modalties and affordances.

surgical_fire 3 hours ago||
When I saw the article I was initially skeptical. I do look down on OpenAI, Google, and other such companies.

But on second thought it is not a bad idea to be able to have a quick tool to identify an image as AI generated.

And after reading your reaction to it, I am sure now that the watermark is for the best.

WhatIsDukkha 3 hours ago||
So you are in the"nothing to hide, nothing to fear" school of privacy rights?

Only criminals and bad actors want private defaults?

The burden of proof is proving there is some harm or problem that needs solving and noone has managed that in this thread or generally.

surgical_fire 38 minutes ago|||
> So you are in the"nothing to hide, nothing to fear" school of privacy rights?

No, but you are in the school that teaches that false equivalence is valid rationale.

> Only criminals and bad actors want private defaults

As I was saying.

> The burden of proof is proving there is some harm or problem that needs solving and noone has managed that in this thread or generally.

"Burden of proof" is a concept borrowed from legal practice where the accuser has to offer proof that the accused commited a crime.

No crime is being implied here. Watermarking is actually a useful feature so that people can easily identify images as AI generated.

dylan604 3 hours ago|||
Not sure what's to hide here. The caveat depends on what data is encoded into the watermark. If it's as simple as the date generated and the system that generated it so that it is easily identifiable as AI generated, I'm fine with it. Hell, I'd even say it'd be cool to embed all of the prompts used to generate the image. If it's also including the name of the user or account ID, then we start getting into gray areas. Since I'm not really on the AI hype train, I'm not all that opposed to that info either. I'll never use it so it won't affect me mindset kicks in on this one, but I'd be okay either way for/against embedding user identifiable info.
janalsncm 5 hours ago|||
Strictly speaking, DRM = digital rights management, which is related to intellectual property.

SynthID would only be DRM if Google/OpenAI were claiming IP rights over their images. I don’t even know if that’s legal though.

WhatIsDukkha 4 hours ago||
What value does "strictly speaking" bring to the discussion?

So that you don't have to address any of the issues?

janalsncm 3 hours ago|||
Words matter? DRM means digital rights management, not simply any kind of metadata a person doesn’t like.
WhatIsDukkha 3 hours ago||
Yes... they do matter, perhaps using care in your understanding before attempting to nitpick?

https://en.wikipedia.org/wiki/Digital_rights_management#Wate...

fc417fc802 1 hour ago||
Perhaps bother to read what you linked before being snarky?

> They are not complete DRM mechanisms in their own right, but are used as part of a system for copyright enforcement ...

Because watermarks in and of themselves are not, in fact, DRM. Even if I agree that their mass adoption by BigTech is a really bad sign for personal privacy and (eventually) freedom.

WhatIsDukkha 1 hour ago||
Yes I did read it years ago and again today?

If you read my original point you'd see I said "weird DRM glorp" which you and other have tried, and failed to only closely parse "DRM" so that you could nitpick poorly.

It is integral and part of DRM systems and certainly "weird DRM glorp" for an actual close reader.

DRM is not just "I cant watch X movie because DRM" even if that is the statistically prevalent understanding of DRM.

Its a suite of technologies of which watermarking is one of.

dylan604 3 hours ago||||
Because DRM is primarily used to ensure the content is not shared in a way the owner does not allow. That is not what SynthID is doing. All it does is allow people to know it is a generated image specifically for when it starts to be widely shared on the internet.

So strictly speaking brings a lot to the discussion when you actually think about it. Stating that DRM != SynthID is addressing issues where people seem to think that DRM == SynthID. Those people are wrong, and strictly speaking need to be corrected.

WhatIsDukkha 3 hours ago||
You are making a category error --

"this image made by OpenAI" is a drm assertion

You wont be able to assert copyright of the picture that you added an OpenAI red bowtie to, thats a DRM issue.

drdeca 3 hours ago|||
Accuracy is valuable.
runarberg 4 hours ago|||
For the record: https://en.wikipedia.org/wiki/Printer_tracking_dots
Barbing 6 hours ago|||
> How old is photoshop and why is it exempt?

How does today’s maximum theoretical disinformation output per minute compare to 2021 Photoshop?

WhatIsDukkha 4 hours ago||
Its 2026... people are deliberately choosing to live in their own realities with no care about objective facts or moral choices.

So weird images are a big problem? No they don't matter at all.

Barbing 4 hours ago|||
Political deepfakes on the mind here more than weird stuff.
WhatIsDukkha 4 hours ago||
"I could stand in the middle of 5th Avenue and shoot somebody and wouldn't lose any voters, ok? It's, like, incredible." — Donald Trump

So what does a deepfake matter?

Barbing 2 hours ago||
Are there other politicians? Do disgruntled employees have bosses? Ex partners?

A national news story in the US tonight, Lyft driver caught faking photos of his messy car. Not the most intelligent fraudster as he left the Gemini logo on the corner of the image.

Providing these four examples in good faith :) also generally I _dislike_ DRM

WhatIsDukkha 57 minutes ago||
Against MILLLIONS of legitimate uses which you don't seem to care to protect?

You should also think about whether, suddenly, courts can now trust images they see because this technology exists?

I think thats not even basically plausible.

surgical_fire 3 hours ago|||
If they don't matter, neither does a watermark.
WhatIsDukkha 3 hours ago||
and you need the watermark to tell you the fish with the mustache is fake...

What image is going to change your worldview so radically that the drm saves you?

edit - to be clear you are watermarking 100,000 fishes with mustaches because of your concern over 1 image that "matters" (and you don't even have an image that matters in mind)

surgical_fire 3 hours ago||
Yes, let's pretend they are adding watermarks because of fishes with mustaches.
doctorpangloss 2 hours ago||
You: "performative nonsense! Arbitrary metadata not of my choosing!"

Also you: well, games go through some kind of distribution, which has plenty of telemetry and metadata. Whether it is App Store with notarization, or Steam or Itch who collect analytics and know a lot about you, or your ISP if you self host your eclectic WebGL game from home. Posting on an iPhone or Android phone, to hacker News which has your email address, on your cell network which has IPv6 globally unique addresses...

"But my choosing!" You'll say. It is extremely performative of you to say, "everything that would make me 200% wrong isn't valid."

I don't know. I really hate these vibes-driven reactions to (checks notes) content attribution. Every accusation is a confession in this frame of mind. How do you not see that?

WhatIsDukkha 2 hours ago||
You are asserting that the existence of metadata in other venues to be proof that this form of watermarking metadata is just fine with you and should be for everyone else because... nope don't see any reason listed here.

I have an IP address so therefore this is all fine?

"Every accusation is a confession" also seems like an insinuation that I have something to hide but you have "nothing to hide, nothing to fear"ie the very generic privacy right fallacy.

As for "vibes driven"... this whole technical "fix" is a result of the reactionary "vibe" of the ai moral panic, your "notes" don't seem to be providing any perspective there?

keyle 44 minutes ago||
Is it like metadata in mp3?

If I take a screenshot of an AI image, will that then be seen as an AI image? Is that 'hidden in the image' or as metadata?

amazingamazing 7 hours ago||
Good. Despite people saying it will be removed, I have seen no reproducible repo demonstrating it.
raincole 7 hours ago||
Stable Diffusion with 10%~15% denoising strength. Done.

I tested the day 1 when Nano Banana Pro was released and it worked. It still works today for Nano Banana 2.

I didn't post this anywhere because I (arrogantly) thought saying it publicly would make the internet worse. But it was pure arrogancy: if I came up with this the first day then of course other millions of programmers did too.

That being said, it'll introduce the typical artifacts from SD models and that might be detected by other methods (or just by zooming in a lot and looking carefully).

vunderba 7 hours ago|||
Yup, OOC a while back I put together a ComfyUI node that took in a NB image and start with the smallest amount of denoise strength using Flux.1 (but works with any model), then run img2img with a synthid check incrementing denoise in a loop until it was defeated.

Never released it, but it was obvious to most people in the SD community that denoising using a diffusion model was a relatively trivial means to beat most steganographic watermarks.

londons_explore 6 hours ago||
Yet is in itself fairly trivial to detect assuming you use some open-weight image model as a base.
zulban 6 hours ago||||
> if I came up with this the first day then of course other millions of programmers did too.

Don't sell yourself short. I'm sure it was only hundreds of thousands.

amazingamazing 7 hours ago|||
Post a repro. I can do that too but then the similarity index is weak. The point is that it it looks indistinguishable then the integrity persists.

In my tests the image looks clearly distinct. In other words, if you can tell the difference then it isn’t a good test.

DonsDiscountGas 5 hours ago|||
Probably a lot easier to use a different model in the first place
dvngnt_ 5 hours ago|||
It will but many people won't as i've seen disinformation that could be detected by synth-id.
post_it_loser 6 hours ago||
[flagged]
userbinator 2 hours ago||
Currently, this article is conveniently right next to it: https://news.ycombinator.com/item?id=48200569
4ashz 5 hours ago||
First they verify whether a picture came from OpenAI, then they'll include subscriber data and geolocation.

Well, they'll finally find out that no one wants to look at AI generated pictures or text. Once they do that, the tool will fail for the public and only work for the government.

Gigachad 2 hours ago|
Seemingly the only use for photo realistic ai generation is deception. We are already seeing AI generated video used in political ads in America.
CSMastermind 7 hours ago||
Aren't these kinds of watermarks easy to remove or distort? Seems like they're only helpful as long as people are relying on them sparingly so it's not worth the effort to circumvent.

If social media platforms started banning images with these watermarks seems like they'd be stripped out overnight.

amazingamazing 7 hours ago||
No, they are very resistant to modification that can be done easily. That being said I doubt it is impossible
janalsncm 5 hours ago|||
Yeah cropping, color shifting, resizing and compression don’t remove it. That said, there’s pretty well known workarounds:

https://github.com/wiltodelta/remove-ai-watermarks

surgical_fire 3 hours ago||
The sort of people that generate AI images that need a watermark are not exactly the kind of people prone for this sort of effort.
snissn 7 hours ago|||
I’m surprised! I guess I’m being naive but I would imagine you could pass an image to an image model without synthid and have it reconstruct the image in a net new way without the markers. I guess I’m wrong? That’s cool if the watermarks are so deeply ingrained that they persist
cephei 7 hours ago||
As I understand it, they modify the image by applying a special Gaussian noise filter which affects each pixel in the image in subtle (possibly not reversible) ways. The detecting service will look for this noise pattern to flag it, so even a part of the image is enough to know it was generated by AI.
vitorgrs 5 hours ago||
Yes, Gemini can actually say how much of the image is AI generated.
Tiberium 7 hours ago|||
I still don't think there's a single GitHub repo that actually removes real SynthID watermarks from Nano Banana 2/NBPro outputs. Most of them are just some research projects that haven't achieved this. The only methods so far I've seen are weird tricks with transparency/overlaying the original image if you're using edits, and also using a diffusion model to regenerate the NB-generated image at low noise levels, but this also modifies the original.
vunderba 7 hours ago||
Right I think that’s why you probably need to start with very low levels of denoising and experiment with different approaches.

Set up as a ComfyUI workflow that does a few things: it tries SDXL, Flux, and a couple of different denoising methods at the lowest possible strength (progressively incrementing) to avoid changing the image too much, while also running a SynthID check each time, and repeating this in a loop until the watermark is essentially gone.

At the same time, you’d probably want to add some kind of threshold based on a perceptual hash aka the maximum perceptual quality difference you’re willing to accept.

programd 7 hours ago|||
Define easily. There is an approach that apparently works and is based on spectral analysis of the images.

https://github.com/aloshdenny/reverse-SynthID

toraway 6 hours ago||
FWIW there are a few people in the issues saying that the tool is giving false negatives and the output image gets flagged by the actual Gemini API as having SynthID. Most recently 3 weeks ago without a response.
Arnt 7 hours ago|||
This one was released a few years ago and still seems unbroken. I'm sure it will be broken at some point, but if you have to wait a year or two from when you make a deepfake until you can post it on Facebook, maybe that's enough. Maybe even a month is enough.
ZeWaka 6 hours ago||
I imagine the technique of having AI recreate the image from scratch based on a very detailed description might work.
raincole 6 hours ago||
That'd not work with today's technology. No open model's prompt adherence is anywhere remotely close to ChatGPT/NanoBanana. 'remotely' here is a funny understatement, as I don't have a strong enough word in my vocabulary to describe how far the open models are behind the closed ones.

Writing a more detailed description does not make the models stick to it more.

vunderba 6 hours ago||
Definitely. I run an entire site built around a series of benchmarks that focus on prompts of increasingly difficult complexity with a focus on adherence, and even the state-of-the-art local models are probably only about thirty percent as good as proprietary models like Gemini 3.1 Flash Image and GPT Image 2.

Comparing Qwen-Image, Flux.2, ZiT, NB2, and gpt-image-2

https://genai-showdown.specr.net/?models=qi,nbp3,f2d,g2,zt

mpetrovich 1 hour ago|
Seems inferior to C2PA, which is actually an open standard: https://contentauthenticity.org/
More comments...