alt.hn

4/9/2026 at 8:10:25 PM

Reverse engineering Gemini's SynthID detection

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

by _tk_

4/9/2026 at 10:26:53 PM

Inserting an undetectable 1-bit watermark into a multi megapixel image is not particularly difficult.

If you assume competence from Google, they probably have two different watermarks. A sloppy one they offer an online oracle for and one they keep in reserve for themselves (and law enforcement requests).

Also given that it's Google we are dealing with here, they probably save every single image generated (or at least its neural hash) and tie it to your account in their database.

by coppsilgold

4/10/2026 at 9:03:23 AM

The dual-watermark theory makes alot of sense for defensive engineering. You always assume your outer layer will be broken and so keep a second layer that isn't publicly testable. Same as defence in depth anywhere else. I'm curious - as new models are being built constantly and they're naturally non-deterministic, do you think it's possible for end users to prove that?

by JoshBlythe

4/9/2026 at 10:14:20 PM

Seems like a very low-quality AI-assisted research repo, and it doesn't even properly test against Google's own SynthID detector. It's not hard at all (with some LLM assistance, for example) to reverse-engineer network requests to be able to do SynthID detection without a browser instance or Gemini access, and then you'd have a ground truth.

by Tiberium

4/9/2026 at 10:38:33 PM

I read a lot of comments on HN that say something is not hard, yet don't provide a POC of their own or link to research they have knowledge of.

I also read a lot of comments on HN that start by attacking the source of the information, such as saying it was AI assisted, instead of the actual merits of the work.

The HN community is becoming curmudgeonly and using AI tooling as the justification.

by ddtaylor

4/9/2026 at 11:34:16 PM

becoming? under most posts that even in passing mention using AI tools there are multiple people raising their noses talking about how much they hate AI use

by love2read

4/10/2026 at 12:01:07 AM

Eh, just the same people that have been killing tech forums and closing posts on stack overflow for like ever.

by pixl97

4/10/2026 at 4:38:53 AM

Im confident i saw the watermark in use today, in nano banana, i copied the image from chrome into slack. the resulting upload was a black square with a red dot. and not the image i had generated.

by stevomacdaddy

4/10/2026 at 5:56:48 AM

I remember experiencing something similar. But then iirc I noticed you can draw on a screenshot and I think i was copying the random dots I made by accidental clicks... You sure it wasnt this?

by funcantor

4/9/2026 at 9:19:28 PM

Ok i get that eventually someone was gonna do this but why would we want to purposely remove one of the only ways of detecting if an image is ai generated or not...?

by khernandezrt

4/10/2026 at 3:28:41 AM

Because an attacker will do that the same thing and without sharing that knowledge good actors are in the dark. It's the same reason we share known security problems, since there will be bad actors that discover the same bugs and use them for much worse.

by ddtaylor

4/9/2026 at 9:25:21 PM

It was always going to be available to some people, but not everyone would know or believe that. Now they will.

by lokar

4/10/2026 at 12:06:22 AM

Much like every other thing in the tech world. He'll, it's why AI will kill us off eventually.

If a system depends on every person on the planet not doing one particular thing or the system breaks, expect the system to break quickly.

This is an especially common trope in software. If someone can make software that does something you consider bad, it will happen. Also it's software. There is no difference between it being available to one person or a million. The moment the software exists and can be copied an unbound number of times.

by pixl97

4/9/2026 at 10:28:49 PM

More likely than not it would be used to deanonymise the author.

So it's a "no" by default.

by subscribed

4/9/2026 at 10:59:45 PM

Fundamentally it's a fuzzy signal and people shouldn't rely on it. The general public does not understand Boolean logic (oh, so the SynthID is not there, therefore this image is real). The sooner AI watermarking faces its deserved farcical demise the better.

Also something about how AI is not special and we haven't added or needed invisible watermarks for other ways media can be manipulated deceptively since time immemorial, but that's less of a practical argument and more of a philosophical one.

by akersten

4/9/2026 at 11:25:54 PM

I’m not very well read on the topic and you seen to take a strong “con” stance. Curious to hear why you think it deserves such a demise

by StarlaAtNight

4/9/2026 at 11:36:06 PM

People think that just because they have a way to prove that an image is AI, their worries of misinformation are solved. Better to acknowledge that wherever you look people will be trying to deceive you even if their content won't have as obvious an indicator as SynthID.

by love2read

4/10/2026 at 3:40:56 AM

Not GP, but I’m pretty “con” too.

Because it’s meaningless for what it’s being marketed for. It’s conceptually inverted. It’s a detector that will detect 100% of the stuff that doesn’t mind being detected, and only the dumbest fraction of stuff that doesn’t want to be detected.

No fault of the extremely smart and capable people who built it. It’s the underlying notion that an imperceptible watermark could survive contact with mass distribution… it gives the futile cat-and-mouse vibes of the DRM era.

Good guys register their guns or whatever, bad guys file off the serial numbers or make their own. Sometimes poorly, but still.

All of which would be fine as one imperfect layer of trust among many (good on Google for doing what they can today). The frustrating/dangerous part is that it seems to be holding itself out as reliable to laypeople (including regulators). Which is how we end up responding to real problems with stupid policy.

People really want to trust “detectors,” even when they know they’re flawed. Already credulous journalists report stuff like “according to LLMDetector.biz, 80% of the student essays were AI-generated.” Jerry Springer built an empire on lie detector tests. British defense contractor ATSC sold literal dowsing rods as “bomb detectors,” and got away with it for a while [2].

It’s backward to “assume it’s not AI-origin unless the detector detects a serial number, since we made the serial number hard to remove.” Instead, if we’re going to “detector” anything, normalize detecting provenance/attestation [e.g. 0]: “maybe it’s an original @alwa work, but she always signs her work, and I don’t see her signature on this one.”

Something without a provable source should be taken with a grain of salt. Make it easy for anyone to sign their work, and get audiences used to looking for that signature as their signal. Then they can decide how much they trust the author.

Do it through an open standards process that preserves room for anyone to play, and you don’t depend on Big Goog’s secret sauce as the arbiter of authenticity.

I hear that sort of thinking is pretty far along, with buy-in from pretty major names in media/photography/etc. The C2PA and CAI are places to look if you’re interested [1].

…and that is why I am “con.”

[0] https://contentcredentials.org/

[1] https://c2pa.org/ , https://contentauthenticity.org/

[2] https://en.wikipedia.org/wiki/ADE_651

by alwa

4/10/2026 at 6:10:51 AM

To pass fake/modified image as genuine?

by drdebug

4/9/2026 at 9:25:41 PM

Uh... you can do this pretty easily since day 1. Just use Stable Diffusion with a low denoising strength. This repo presents an even less destructive way[0], but it has always been very easy to hide that an image is generated by Nano Banana.

[0]: if it does what it claims to do. I didn't verify. Given how much AI writing in the README my hunch is that this doesn't work better than simple denoising.

by raincole

4/9/2026 at 9:22:54 PM

SynthID is visible in some generations (areas with a lot of edges, or text), I wonder if this would make them look better.

by M4v3R

4/10/2026 at 10:57:58 AM

And it gets more and more visible in every edit if you use Nano Banana for the edits.

by thecupisblue

4/9/2026 at 9:12:02 PM

kinda ironic you can clearly see signs of Claude, as it shows misaligning table walls in the readme doc

by armanj

4/9/2026 at 9:20:12 PM

Parenthesized, comma-separated lists with no “and” is an even stronger tell. Claude loves those.

by rafram

4/10/2026 at 4:47:09 AM

I also use those extensively, they just flow better, especially if you have an "and" in the surrounding sentence.

by LiamPowell

4/9/2026 at 9:58:07 PM

> kinda ironic you can clearly see signs of Claude, as it shows misaligning table walls in the readme doc

This one is such a gigantic clusterfuck... They're mimicking ASCII tables using Unicode chars of varying length and, at times, there's also an off-by-one error. But the model (not Claude, but the model underneath it) is capable of generating ASCII tables.

P.S: I saw the future... The year is 2037 and we've got Unicode tables still not properly aligned.

by TacticalCoder

4/9/2026 at 10:09:23 PM

I mean, just reading the readme content it is pretty obvious it is Claude

by dgellow

4/9/2026 at 10:10:08 PM

[dead]

by huflungdung

4/9/2026 at 10:09:29 PM

Okay... this tests its own ability to remove the watermark against its own detector. It doesn't test against Gemini's SynthID app. So it does nothing...

by doctorpangloss

4/9/2026 at 8:55:57 PM

It says not to use these tools to misrepresent AI-generated content as human-created. But the project is a watermark removal tool with a pip-installable CLI and strength settings named "aggressive" and "maximum." Calling this research while shipping turnkey watermark stripping is trying to have it both ways in a way that's uncomfortable to read.

The README itself reads like unedited AI output with several layers of history baked in.

- V1 and V2 appear in tables and diagrams but are never explained. V3 gets a pipeline diagram that hand-waves its fallback path.

- The same information is restated three times across Overview, Architecture, and Technical Deep Dive. ~1600 words padded to feel like a paper without the rigor.

- Five badges, 4 made up, for a project with 88 test images, no CI, and no test suite. "Detection Rate: 90%" has no methodology behind it. "License: Research" links nowhere and isn't a license.

- No before/after images, anywhere, for a project whose core claim is imperceptible modification.

- Code examples use two different import styles. One will throw an ImportError.

- No versioning. If Google changes SynthID tomorrow, nothing tells you the codebook is stale.

The underlying observations about resolution-dependent carriers and cross-image phase consistency are interesting. The packaging undermines them.

by refulgentis

4/9/2026 at 9:15:21 PM

Agreed. This isn't punk this just helps the bad guys. Society needs to know what content is AI generated and what is not.

by jonshariat

4/9/2026 at 9:41:17 PM

This was never going to be a reliable way to do it. It's basically the evil bit . It only works for as long as everyone is making a good-faith effort to follow the convention. But the bad guys do not do that.

by recursive

4/9/2026 at 9:19:20 PM

If that's the case, society will inevitably be disappointed.

There are already ten million AI image generators, the overwhelming majority of which do not watermark their outputs. Google auto-inserting them is nice, but ultimately this kind of tool to remove them will inevitably be widespread.

by SR2Z

4/9/2026 at 9:49:22 PM

It really doesn't need such capability. Nor does it need the capability to know what human generated it either.

by charcircuit

4/9/2026 at 9:09:34 PM

if you downscale then upscale it removes the watermark

by kelsey98765431

4/9/2026 at 9:51:38 PM

I don't understand all the handwringing. If it's this easy to remove SynthID from an AI-generated image then it wasn't a good solution in the first place.

by sodacanner

4/9/2026 at 9:54:12 PM

There is no solution. I don't know why people discuss this subject as if there is a technical solution. As if there are fairies or souls hidden in the pixels that help us tell what is AI generated and what is not.

by raincole

4/9/2026 at 10:05:25 PM

If you want to make an AI generated image but don't want other people to know that it's AI, the most obvious solution is to not use Gemini. Synth ID is watermarking. It's only ever going to be useful to good actors, who want an AI generated image and aren't trying to hide the fact that it's AI generated.

by DonsDiscountGas

4/9/2026 at 10:42:20 PM

Never underestimate that people are lazy.

by dummydummy1234

4/9/2026 at 9:58:47 PM

Sure, and things like this help drive home that SynthID wasn't a solution at all.

by sodacanner

4/9/2026 at 10:18:11 PM

Sure there is a solution, you are just looking at it the wrong way. Make non-AI images provably unaltered with signed keys from the device (e.g. the camera) that took it.

by levocardia

4/9/2026 at 10:56:29 PM

That's pretty much impossible though.

One workflow that some artists use is that they draw with ink on paper, scan, and then digitally color. Nothing prevents someone from generating line art using generative AI, printing it, scanning it, and coloring it.

And what if someone just copy pastes something into Photoshop or imports layers? That's what you'd do for composites that mix multiple images together. Can one copy paste screenshots into a multi layer composition or is that verboten and taints the final image?

And what about multi program workflows? Let's say I import a photo, denoise it in DxO, retouch in affinity photo, resize programmatically using image magick, and use pngcrush to optimize it, what metadata is left at the end?

by jfim

4/10/2026 at 12:09:16 AM

Next comes registration your camera with the government to ensure you're not doing "bad" things with it.

by pixl97

4/9/2026 at 10:53:05 PM

If the premise is that everyone would just agree on the same protocol, I have an even more unbreakable solution: every image has to be upload to a blockchain the moment it is (claimed to be) created. Otherwise it's AI.

If only everyone just agrees with me.

by raincole

4/9/2026 at 10:22:16 PM

Which works for about 5 minutes until someone leaks a manufacturer's private key or extracts it from a device...

by Diggsey

4/9/2026 at 10:25:29 PM

How many minutes do you think it would take before someone figured out how to crack that?

by IncreasePosts

4/9/2026 at 10:36:26 PM

On Pixels and iPhones it would be impossible since they have actually secure hardware that could both hold the keys and sign/verify the image.

by subscribed

4/9/2026 at 10:45:46 PM

The camera module sits outside the secure area, meaning it would need to send data in to be signed. How does the phone know that it's getting legitimate data from the camera module, or data someone else is just piping in? Also, you could probably get a fairly high quality image by just taking a photo of something AI generated in the right lighting conditions.

by IncreasePosts

4/9/2026 at 9:54:02 PM

Yes. This kind of project needs aggressive red teaming, it leads to better products and we need excellent products in this space.

This project proves what red teaming was in place wasn't good enough.

by rustyhancock

4/10/2026 at 11:53:44 AM

[dead]

by enesz

4/10/2026 at 9:13:47 AM

[dead]

by neuzhou

4/10/2026 at 7:01:19 AM

[dead]

by jiusanzhou

4/10/2026 at 12:15:31 AM

[dead]

by techpulselab

4/10/2026 at 12:07:17 AM

[dead]

by matthias_m_dev

4/9/2026 at 8:51:27 PM

> We're actively collecting pure black and pure white images generated by Nano Banana Pro to improve multi-resolution watermark extraction.

Oh hey, neat. I mentioned this specific method of extracting SynthID a while back.[1]

Glad to see someone take it up.

[1]: https://news.ycombinator.com/item?id=47169146#47169767

by andrewmcwatters

4/9/2026 at 9:07:54 PM

FWIW, I had Nano Banana create pure white/black images in February, and there was no recognizable watermark in them (all pixels really were #ffffff / #000000 IIRC).

Meta: your comment was marked [dead], like a few other constructive comments I saw in recent days. Not sure why.

by raphman

4/10/2026 at 2:05:50 AM

I tried it with Nano Banana 2 through the API just now, and it was content filtering me on both white and black images.

by pattilupone

4/9/2026 at 9:36:41 PM

I suspect they strip the SynthID for these specific cases to prevent exfiltration of the steganography.

I appreciate you pointing it out, but this account is banned. Thank you for vouching though!

by andrewmcwatters