Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Apple defends anti-child abuse imagery tech after claims of ‘hash collisions’

vice.com

Post reply on HN
Print view View original

191–200 of 465 posts

Re: Apple defends anti-child abuse imagery tech after claims of ‘hash collisions’

#191

Post by Ashanmaril »

ALittleLight wrote: post #102

Would this work as an attack? 1. Get a pornographic picture involving young though legal actors and actresses. 2. Encode a nonce into the image. Hash it checking for CSAM collisions. If you've found a collision go on to the next step, if not update the nonce and try again. 3. You now have an image that, to visual inspection will appear plausibly like CSAM, and to automated detection will appear like CSAM. Though, pre…

If you're gonna go through that much effort, you might as well just text someone an actually illegal image from something other than an iPhone

You're already attempting to frame someone for a crime, might as well commit another crime while you're at it

Re: Apple defends anti-child abuse imagery tech after claims of ‘hash collisions’

#192

Post by pasquinelli »

GeekyBear wrote: post #175
mightybyte wrote: post #162

Earlier quoted context omitted.

I think the difference here is that it's ON YOUR DEVICE. I think there's a pretty clear understanding that if you upload stuff to a cloud provider they can do whatever they want with it. This is different. This is reaching into what has up until now mostly been considered a private place. Law enforcement often has to get warrants to search this kind of thing. This is the difference between putting CSAM on a sign in y…

ON YOUR DEVICE (where it's encrypted in a way that Apple can't read until the 30 image threshold is crossed) is more private than doing the same scan on server where a single false positive can be misused by anyone who can get a subpoena.

what kind of encryption can't be decrypted until some threshold has been crossed?

maybe you mean to say that apple says they won't read it until that threshold has been crossed.

Re: Apple defends anti-child abuse imagery tech after claims of ‘hash collisions’

#193

Post by laverya »

floatingatoll wrote: post #56
smoldesu wrote: post #35

Earlier quoted context omitted.

Furthermore, it may well be possible to combine that blobby grey static with another image, manipulating it's visual hash to create a "sleeper" positive. If this was possible in a week , then it's going to be very interesting to watch the technology change/evolve over the next few years.

Doing so would create a positive that still doesn't pass Apple's human visual check against the (blurred) CSAM content associated with that checksum, and if it somehow did, it would still then also have to occur at qty.30 or more, and they'd have to pass a human visual check against the (unblurred) CSAM content by one of the agencies in possession of it. It's not possible to spoof that final test unless you possess r…

And if you make the second image be actual, 18+ porn? Will Apple be able to tell the difference between that and CSAM after the blur is applied?

Bonus points if you match poses, coloration, background, etc.

Re: Apple defends anti-child abuse imagery tech after claims of ‘hash collisions’

#194

Post by robertoandred »

ALittleLight wrote: post #102

Would this work as an attack? 1. Get a pornographic picture involving young though legal actors and actresses. 2. Encode a nonce into the image. Hash it checking for CSAM collisions. If you've found a collision go on to the next step, if not update the nonce and try again. 3. You now have an image that, to visual inspection will appear plausibly like CSAM, and to automated detection will appear like CSAM. Though, pre…

Why would you save to your personal photo library some porn that a creepy rando sent to you in a text?

Re: Apple defends anti-child abuse imagery tech after claims of ‘hash collisions’

#195

Post by fraa-orolo »

__blockcipher__ wrote: post #183
fraa-orolo wrote: post #171

Earlier quoted context omitted.

Because it's not a cryptographic hash where a one bit difference results in a completely different hash. It's a perceptual hash that operates on a smaller bitmap derived from the image so it's plausible that some innocuous images might result in similar derivations; and there might be intentionally crafted innocently-looking images that result in an offensive derivative. Salvador Dali could do something similar by ha…

This is a great answer but that’s not actually the GP’s contention. Their argument is essentially “so what if there’s a collision, the human review will catch it”. And to that I’d say that the same is supposed to occur for the no-fly list and we all know how that works in practice. The mere accusal itself of possessing CSAM can be life ruining if it gets to that stage. More importantly, a collision will effectively a…

Indeed, I touched on that in another comment: https://news.ycombinator.com/item?id=28227141

Re: Apple defends anti-child abuse imagery tech after claims of ‘hash collisions’

#196

Post by salawat »

rootusrootus wrote: post #51
salawat wrote: post #34

Earlier quoted context omitted.

Take 2 CSAM pictures, known. Combine into one image. Swing and a miss. Not in the CSAM dataset. Take two images. Encode alternating pixels. Decode to get the original image back. Convert to different encodings print to PDF or Postscript. Encode as base64 representations of the image file... Who are we trying to fool here? This is kiddie stuff. This is more about trying to implant scanning capabilities on client devic…

> This is more about trying to implant scanning capabilities on client devices. Did we think they didn't already have that ability?

No. The brazen part of this is Apple trying to sell the public on this being okay, effective to the point of mitigating the abuse potential. False negative AND false positive have to counterweight the existence of the vector for abuse. That's what you're measuring against. You can't even guarantee that. This isn't even close to a safe or effective move. The risk management on this is hosed.

Re: Apple defends anti-child abuse imagery tech after claims of ‘hash collisions’

#197

Post by mavhc »

_2d30 wrote: post #100
endisneigh wrote: post #52

Am I the only one who finds no issue with this? Personally I’d prefer this than no encryption and scanning in the cloud, which is already possible. All of the slippery slope arguments have already been possible for nearly a decade now with the cloud. Can someone illustrate something wrong with this that’s not already possible today. Fundamentally unless you audited the client and the server yourself either (client or…

Why prefer your owned device tattling on you to Apple looking at data you give them on their servers? The reason people don't like this, as opposed to, for example, Dropbox scanning your synced files on their servers, is that a compute tool you ostensibly own is now turned completely against you. Today, that is for CSAM, tomorrow, what else?

Why did you think you owned a closed source device?

Re: Apple defends anti-child abuse imagery tech after claims of ‘hash collisions’

#198

Post by balozi »

They are slowly turning what is a philosophical argument into a technical question. Next step is for them to conjure up a technical answer before claiming victory. Most people are already bored by all the technical talk.

Neural hash this: Its about Trust. Its about Privacy. Its about Boundaries between me and corporations/governments/etc.

Re: Apple defends anti-child abuse imagery tech after claims of ‘hash collisions’

#199

Post by shapefrog »

petersellers wrote: post #76
endisneigh wrote: post #52

Am I the only one who finds no issue with this? Personally I’d prefer this than no encryption and scanning in the cloud, which is already possible. All of the slippery slope arguments have already been possible for nearly a decade now with the cloud. Can someone illustrate something wrong with this that’s not already possible today. Fundamentally unless you audited the client and the server yourself either (client or…

Personally, I'd prefer no privacy intrusions at all. The issue is that Apple previously was not intruding into their user's privacy (at least publicly), but now they are. It sounds like your argument is that Apple could have been doing this all along and just not telling us. I find that unlikely mainly because they've marketed themselves as a privacy-focused company up until now.

> The issue is that Apple previously was not intruding into their user's privacy

Apple reserves the right at all times to determine whether Content is appropriate and in compliance with this Agreement, and may screen, move, refuse, modify and/or remove Content at any time, without prior notice and in its sole discretion, if such Content is found to be in violation of this Agreement or is otherwise objectionable.

You must not have been paying attention for the last 20 years.

Re: Apple defends anti-child abuse imagery tech after claims of ‘hash collisions’

#200

Post by codeecan »

slg wrote: post #149
carnitas wrote: post #145

Earlier quoted context omitted.

Apple inspects every file on the local device Before its uploaded. It’s just pinky promise only matched with the on device database when an upload is intended.

Apple controls the hardware, software, and cloud service. It was always a pinky promise that they wouldn't look at your files. I don't know why we should doubt that pinky promise less today than we did a month ago.

They don't control the database used, any country can thru legal means attach additional hashes for search and reporting.

Apple has already proven it will concede to China's demands.

They are building the worlds most pervasive surveillance system and when the worlds governments come knocking to use it ... they will throw their hands up and feed you the "Apple complies with all local laws etc.."

Post reply on HN