The White House wants to ‘cryptographically verify’ videos of Joe Biden so viewers don’t mistake them for AI deepfakes::Biden’s AI advisor Ben Buchanan said a method of clearly verifying White House releases is “in the works.”
The White House wants to ‘cryptographically verify’ videos of Joe Biden so viewers don’t mistake them for AI deepfakes::Biden’s AI advisor Ben Buchanan said a method of clearly verifying White House releases is “in the works.”
Apple’s scrapped on-device CSAM scanning was based on perceptual hashes.
The first collision demo breaking them showed up in hours with images that looked glitched. After just a week the newest demos produced flawless images with collisions against known perceptual hash values.
In theory you could create some ML-ish compact learning algorithm and use the compressed model as a perceptual hash, but I’m not convinced this can be secure enough unless it’s allowed to be large enough, as in some % of the original’s file size.
you can definitely produced perceptual hashes that collide, but really you’re not just talking about a collision, you’re talking about a collision that’s also useful in subverting an election, AND that’s been generated using ML which is something that’s still kinda shakey to start with
Perceptual hash collision generators can take arbitrary images and tweak them in invisible ways to make them collide with whichever hash value you want.
from the comment above, it seems like it took a week for a single image/frame though… it’s possible sure but so is a collision in a regular hash function… at some point it just becomes too expensive to be worth it, AND the phash here isn’t being used as security because the security is that the original was posted on some source of truth site (eg the whitehouse)
No, it took a week to refine the attack algorithm, the collision generation itself is fast
The point of perceptual hashes is to let you check if two things are similar enough after transformations like scaling and reencoding, so you can’t rely on that here
oh yup that’s a very fair point then! you certainly wouldn’t use it for security in that case, however there are a lot of ways to implement this that don’t rely on the security of the hash function, but just uses it (for example) to point to somewhere in a trusted source to manually validate that they’re the same
we already have the trust frameworks; that’s unnecessary… we just need to automatically validate (or at least provide automatic verifyability) that a video posted on some 3rd party - probably friendly or at least cooperative - platform represents reality