Apple's child protection features spark concern within its own ranks: sources • Hacker News Board

Re: Apple's child protection features spark concern within its own ranks: sources

#761

Post by zepto » Fri, Aug 13, 2021, 6:25 PM UTC

Aaargh20318 wrote: post #757

zepto wrote: post #717

Earlier quoted context omitted.

> And when ‘manually’ checking, they aren’t actually looking at potential CSAM, they are checking the ‘visual derivative’ (whatever that means exactly), basically a manual check if the hashes match. The visual derivative is enough to tell the difference between a collection of CSAM and say documents, or pictures of protests. It can’t be abused without Apple’s involvement.

> The visual derivative is enough to tell the difference between a collection of CSAM and say documents, or pictures of protests. How would you know ? They have never given details on what this 'visual derivative' actually is. If it's detailed enough to recognise it as CSAM, then Apple isn't allowed to look at it.

That’s a fair point.

However this is moot anyway. Using this mechanism to search for things other than CSAM would be unconstitutional under the 4th amendment.

https://www.economist.com/united-states/2021/08/12/a-38-year...

Re: Apple's child protection features spark concern within its own ranks: sources

#762

Post by zepto » Fri, Aug 13, 2021, 6:26 PM UTC

fsflover wrote: post #671

zepto wrote: post #198

Earlier quoted context omitted.

This is the most honest take on this that I’ve seen. The slippery slope arguments don’t make sense, nor does the risk of false positives. But the idea of being suspected even in this abstract way, because of something other people do , is at the very least distasteful, bordering on offensive.

> The slippery slope arguments don’t make sense Yes, they do: https://news.ycombinator.com/item?id=28160092

No they don’t. Searches for things other than CSAM would be inadmissible under the 4th amendment. https://www.economist.com/united-states/2021/08/12/a-38-year...

Re: Apple's child protection features spark concern within its own ranks: sources

#763

Post by dukeofdoom » Fri, Aug 13, 2021, 6:30 PM UTC

hn_throwaway_99 wrote: post #223

whoisjuan wrote: post #132

This CSAM Prevention initiative by Apple is a 180 degress change of their general message around privacy. Imagine investing hundreds of millions of dollars in pro-privacy programs, privacy features, privacy marketing, etc... just to pull this reverse card. Of course this is going to spark concern within their own ranks. It's like working for a food company that claims to use organic, non-processed, fair-trade ingredi…

I actually think something else happened, and to be honest I think many at Apple behind this decision are likely pretty surprised by the blowback. That is, it seems like Apple really wanted to preserve "end-to-end" encryption, but they needed to do something to address the CSAM issue lest governments come down on them hard. Thus, my guess is, at least at the beginning, they saw this as a strong win for privacy. As th…

Apple know what they are doing. Don't fool yourself by giving plausible excuses. The further away you are from your phone the more privacy you will have.

Almost everyone was traced back to the Jan 6th insurrection, because unlike antifa they took their phones with them, so easy trace for government. Now they are in solitary confinement in prison while the world loughs at their expectations of privacy from their phones.

If they get a covid passport implemented, you won't be able to go out in public without your phone to prove you've been vaccinated. So leaving a phone behind will not be much of an option if this is to pass. Metadata and contact tracing will keep track of everyone you're near. This dystopian future is possibly just a few weeks away.

Re: Apple's child protection features spark concern within its own ranks: sources

#764

Post by maerF0x0 » Fri, Aug 13, 2021, 6:30 PM UTC

I maintain that I actually think Apple got told do this by the Gov't the same way AT&T was told -- and gagged- to spy on americans.

Re: Apple's child protection features spark concern within its own ranks: sources

#765

Post by zepto » Fri, Aug 13, 2021, 6:32 PM UTC

mkmk wrote: post #642

zepto wrote: post #486

Earlier quoted context omitted.

Neither 1 nor 2 are in fact true. Spotlight indexes all kinds of metadata, as does photos search. Adding an agent to upload data from these is easier than extending the CSAM mechanism, and the CSAM mechanism as is is not all that plausible to abuse either technically or socially given how clear Apple’s promises are.

Apple very recently promised that “what happens on iPhone stays on iPhone”. At literally the same time, they were building a system that was designed to break that promise. Why should we believe their new promises when their last promise was disingenuous? https://9to5mac.com/2019/01/05/apple-privacy-billboard-vegas...

How does this break that promise? Nothing leaves the phone unless the user chooses to upload their photos to a cloud service.

Re: Apple's child protection features spark concern within its own ranks: sources

#766

Post by psyc » Fri, Aug 13, 2021, 6:32 PM UTC

zepto wrote: post #762

fsflover wrote: post #671

Earlier quoted context omitted.

> The slippery slope arguments don’t make sense Yes, they do: https://news.ycombinator.com/item?id=28160092

No they don’t. Searches for things other than CSAM would be inadmissible under the 4th amendment. https://www.economist.com/united-states/2021/08/12/a-38-year...

That article is very obviously not about what is being searched for. It's about who it is being reported to. For example, if Apple one day decided to detect unauthorized copyrighted media, they could report it to the RIAA. The RIAA is independent of the government, thus the 4th amendment is not violated.

Re: Apple's child protection features spark concern within its own ranks: sources

#767

Post by ribosometronome » Fri, Aug 13, 2021, 6:35 PM UTC

alwillis wrote: post #493

ribosometronome wrote: post #390

Earlier quoted context omitted.

Why do you think essentially no one is complaining about using ML to understand the content of photos, then (especially in comparison to this rather targeted CSAM feature)? My impression is that both Apple and Google have already been doing that since what? 2016? Earlier? There's been no need for a database of photos, either company could silently update those algorithms to ping on guns, drugs, Winnie the Pooh memes,…

Why do you think essentially no one is complaining about using ML to understand the content of photos… At last in Apple's case, ML is used only if a minor child (less than 13 years old) who is on a Family account where the parent/guardian has opted-in to the ability to be alerted if potentially bad content is either sent or received using the Messages app.

Google and Apple both use ML to understand the content of photos already. Go into your app and you can search for things like "dog", "beer", "birdhouse", and "revolver". Given that it can do all that (and already knows a difference between type of guns), it doesn't seem like a stretch to think it could, if Apple or Google wanted, understand "cocaine", "Winnie the Pooh memes", or whatever repressive-government style worries are had. And it's existed for years, without slippery sloping into anything other than this Messages feature for children.

Re: Apple's child protection features spark concern within its own ranks: sources

#768

Post by nextlevelwizard » Fri, Aug 13, 2021, 6:36 PM UTC

goldenkey wrote: post #758

nextlevelwizard wrote: post #753

Earlier quoted context omitted.

>You should avoid making political or moral arguments I am not being political or moral. Why would you think that? Is this one of those "since you don't immediately agree with me it means you are the enemy and disagree on everything" type things? >An image is orders of magnitude larger than a password, meaning that many more collisions. Whole point of hashing algorithms is that they produce unique hashes for differen…

> Whole point of hashing algorithms is that they produce unique hashes for different inputs. Wouldn't a larger input value (bits of the image) make it harder to have collisions? Standard cryptographic hashes should have certain properties like https://en.wikipedia.org/wiki/Avalanche_effect However, a perceptual hash desires the opposite. Due to the pigeonhole principle, a larger space mapping to a smaller space, invo…

Thanks for that, I still dont get why this is such a big privacy concern for everyone.

Re: Apple's child protection features spark concern within its own ranks: sources

#769

Post by shishy » Fri, Aug 13, 2021, 6:41 PM UTC

cucumb3rrelish wrote: post #680

pdabbadabba wrote: post #676

Earlier quoted context omitted.

Please look more carefully at that document. It says, on page 5: "Instead of scanning images in the cloud, the system performs on-device matching using a database of known CSAM image hashes provided by NCMEC and other child safety organizations." And look at the diagram. The matching of image hashes with user photos occurs "on device."

On device != Scanning your device. It scans your iCloud uploads and if configured, iMessages (which you didnt mean, because that doesnt work with hashes)

The general worry is that the fact that it happens on the device means in the future it could conceivably scan your device, even if there are some software checks in place to only scan things being synced to iCloud.

Re: Apple's child protection features spark concern within its own ranks: sources

#770

Post by throwaway93832 » Fri, Aug 13, 2021, 6:45 PM UTC

rdiddly wrote: post #513

Can we talk about the irony that the NCMEC's database of CSAM (two abbreviations I didn't know before this week) is literally a huge child porn collection, or is that too immature? The whole story sounds like bad science fiction to me, I gotta say. But sadly I think we have to beware of anything that's packaged as anti-child-porn just like whatever's anti-terrorism, anti-crime etc. etc. You should be thinking "What i…

Only the priests of high are sacred enough to handle the artifices, to administer their truths to the gentiles.

https://www.latimes.com/california/story/2021-05-05/lapd-off...