Live data from Hacker News

Researchers tested AI watermarks and broke all of them

wired.com

31–40 of 91 posts

Re: Researchers tested AI watermarks and broke all of them

#31
post #4

People have been trying to watermark digital media for decades, when there was (still is) a very strong financial incentive to get it working. It never worked. I don’t think it ever will work.

You are confusing access restrictions with signing. You can easily sign digital media to show that it was made by you.

Re: Researchers tested AI watermarks and broke all of them

#32

Earlier quoted context omitted.

I think you're essentially describing the hardware DRM supply chain. For example, HDCP is a DRM scheme where Intel convinces (or legally requires) every manufacturer of HDMI output devices (e.g. set-top boxes, Blu-ray players) in the world to encrypt certain video streams. Then, Intel requires manufacturers of HDMI input devices (e.g. TVs) to purchase a license key that can decrypt those video streams. This license a…

Interesting. I don't understand the revocation process though. What stops the blu-ray reader from just ignoring the revocation list on the disk?

That's where the reversing comes in to switch the function call to check the revocation list to a NOP and just keep on going. At least, that's how I imagine HDMI equipment that ignores HDCP works

What stops them from being sold that way would probably be the licensing agreement and honest players. I'd imagine in China, there are lots of these types of devices available.

Re: Researchers tested AI watermarks and broke all of them

#33
post #4

People have been trying to watermark digital media for decades, when there was (still is) a very strong financial incentive to get it working. It never worked. I don’t think it ever will work.

You are confusing access restrictions with signing. You can easily sign digital media to show that it was made by you.

You are confusing a digital signature for evidence of anything other than an attestation.

If you create a digital record, then sign it, then that signature is only an attestation of may claim you make, not evidence of that claim. That is the problem with relying on technology to establish trust - the moment you attach an economic benefit to a technology you incentivize people to circumvent it, or to leverage it to commit fraud.

Re: Researchers tested AI watermarks and broke all of them

#34

Earlier quoted context omitted.

I think you're essentially describing the hardware DRM supply chain. For example, HDCP is a DRM scheme where Intel convinces (or legally requires) every manufacturer of HDMI output devices (e.g. set-top boxes, Blu-ray players) in the world to encrypt certain video streams. Then, Intel requires manufacturers of HDMI input devices (e.g. TVs) to purchase a license key that can decrypt those video streams. This license a…

Interesting. I don't understand the revocation process though. What stops the blu-ray reader from just ignoring the revocation list on the disk?

the blu-ray is encrypted and the player requires a key to decrypt it

if you want your blu-ray player to be able to read blu-ray disks you sign a contract that says you will respect the revocation list

if you change your mind later: your player key will be revoked and new blu-rays won't play on your device

(it's actually more sophisticated than this... they can block specific players too)

Re: Researchers tested AI watermarks and broke all of them

#35

Earlier quoted context omitted.

I think you're essentially describing the hardware DRM supply chain. For example, HDCP is a DRM scheme where Intel convinces (or legally requires) every manufacturer of HDMI output devices (e.g. set-top boxes, Blu-ray players) in the world to encrypt certain video streams. Then, Intel requires manufacturers of HDMI input devices (e.g. TVs) to purchase a license key that can decrypt those video streams. This license a…

>Of course, Intel's HDCP master key was either leaked or reverse-engineered, so anyone can generate their own valid device keys Of an older version of HDCP. New media can require a higher HDCP version where that bypass isn't possible.

That's certainly possible, but did that actually happen in practice? The downside is obviously that you essentially start selling new Blu-ray discs that don't work on any old Blu-ray players. I feel like I would have heard if that happened. Unless maybe the old players could issue firmware updates?

Re: Researchers tested AI watermarks and broke all of them

#36
post #17
post #10

We need to focus on the other direction. How can we have chains of trust for content creation, such as for real video. Content can be faked, but not necessarily easily faked from the same sources that make use of cryptographic signing. The attacks can sign the own work, so you'd need ways to distinguish those cases, but device level keys, organizational keys, distribution keys all can provide provenance chains that c…

I was thinking the other day about embedding keys in cameras, etc. but came up with the problem that you could just wire up a computer that BEHAVES like a CCD sensor and send whatever the hell you feel like in to the signing hardware, so you feed in your fake image and it gets signed by the camera as though it were real. I assume smarter people than me have put much more time into the problem, so I'd be interested to…

Every time I hear about these watermarks that's what I think. There is nothing stopping someone from playing any video on a projector and simply recording the projected video on any device they want.

Re: Researchers tested AI watermarks and broke all of them

#37

Earlier quoted context omitted.

I think you're essentially describing the hardware DRM supply chain. For example, HDCP is a DRM scheme where Intel convinces (or legally requires) every manufacturer of HDMI output devices (e.g. set-top boxes, Blu-ray players) in the world to encrypt certain video streams. Then, Intel requires manufacturers of HDMI input devices (e.g. TVs) to purchase a license key that can decrypt those video streams. This license a…

Interesting. I don't understand the revocation process though. What stops the blu-ray reader from just ignoring the revocation list on the disk?

The revocation list is for the TV. Intel revokes a TV's key, distributes the updated revocation list on new Blu-ray discs, and when a compliant Blu-ray player is playing one of those new discs it will refuse to negotiate with a revoked TV.

Now that I think of it, I wonder if compliant Blu-ray players actually save the new revocation entries and then continue refusing to negotiate with revoked TVs even for old Blu-ray discs.

Re: Researchers tested AI watermarks and broke all of them

#38
post #34

Earlier quoted context omitted.

Interesting. I don't understand the revocation process though. What stops the blu-ray reader from just ignoring the revocation list on the disk?

the blu-ray is encrypted and the player requires a key to decrypt it if you want your blu-ray player to be able to read blu-ray disks you sign a contract that says you will respect the revocation list if you change your mind later: your player key will be revoked and new blu-rays won't play on your device (it's actually more sophisticated than this... they can block specific players too)

That doesn't quite sound right to me. I don't think a new Blu-ray disc could be released that continues to be readable by some old readers but is no longer readable by other old readers.

Re: Researchers tested AI watermarks and broke all of them

#39
post #10

We need to focus on the other direction. How can we have chains of trust for content creation, such as for real video. Content can be faked, but not necessarily easily faked from the same sources that make use of cryptographic signing. The attacks can sign the own work, so you'd need ways to distinguish those cases, but device level keys, organizational keys, distribution keys all can provide provenance chains that c…

I was thinking the same thing- we've had digital signing algorithms for decades, and those seem to work fine enough. There's a healthy distrust of cert authorities, sure, but it still works.
Post reply on HN