back

by kristjansson·5y ago·view on hn ↗
It's not just that they're hard to get - it's also that no one has demonstrated a preimage attack. So given just an illegal hash one can't construct an innocent image with that hash. These demonstrations show only that a second image can be transformed to have the same hash as a first, given image.

As other comments note, that's not a huge increase in difficulty for an adversary willing to deal in actual CSAM.

1 comments
> no one has demonstrated a preimage attack. So given just an illegal hash one can't construct an innocent image with that hash. These demonstrations show only that a second image can be transformed to have the same hash as a first, given image.

Isn’t this exactly like that? Or am I missunderstanding something.

https://twitter.com/ghidraninja/status/1428269674912002048?s...

No. The author describes their process for generating the image down that very thread[0]

Quoting the tweet for those who dislike twitter:

> The procedure for this was pretty simple:

> - Started with a placeholder image

> - Got the hash of it

> - Changed text of the image to match the hash

> - Then used

> @anishathalye great neural-hash-collider which took only 1-2 iterations

N.B. it starts with 'started with an image, and the hash of it'

[0]: https://twitter.com/ghidraninja/status/1428270675010199554

What is the practical difference? It is forged innocent image which has matching hash. You basically started with only a hash, as image was not related to it.