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

Apple plans to scan US iPhones for child abuse imagery

ft.com

Post reply on HN
Print view View original

361–370 of 390 posts

Re: Apple plans to scan US iPhones for child abuse imagery

#361

Post by headShrinker »

babyshake wrote: post #173
mm983 wrote: post #161

Earlier quoted context omitted.

This is painful to read with the arrogant undertone while you don't actually know how CSAM scanning works. It's fuzzy hash, not ""AI"", based. Cloudflare uses it too, and last time i checked the web is still functional. https://support.cloudflare.com/hc/en-us/articles/36004610611... How about we turn the tables and have the complainers suggest a solution. Because every single time an approach to targeted child porn t…

So this means it is checking if you are sharing known CP images? That does seem to be much less invasive and problematic as there is likely no good reason to be sharing these images.

> So this means it is checking if you are sharing known CP images?

No. NeuralMatch was “trained” using 200,000 CP images. “Neural“ is likely a reference to the perceptual matching that it uses. It is not a bit for bit match.

Perceptual matching is a technique used for categorizing images based on characteristics and content.

The algorithm will scan your library containing new information and compare it to what it understands as CP.

Re: Apple plans to scan US iPhones for child abuse imagery

#362

Post by headShrinker »

mm983 wrote: post #161
cronix wrote: post #124

Earlier quoted context omitted.

Don't worry, AI is very good at these things and never makes mistakes. It is so advanced it even understands nuance and complex communication mechanisms like sarcasm in text, even if it's only a single sentence to contrast! It will be able to know exactly the context of each image and it will affect only the sickos. We're also working on watching your face through the camera to see your true expression when viewing t…

This is painful to read with the arrogant undertone while you don't actually know how CSAM scanning works. It's fuzzy hash, not ""AI"", based. Cloudflare uses it too, and last time i checked the web is still functional. https://support.cloudflare.com/hc/en-us/articles/36004610611... How about we turn the tables and have the complainers suggest a solution. Because every single time an approach to targeted child porn t…

There is no reference to fuzzy anything in any of the write ups. I don’t know how you can proclaim that this is a fuzzy hash. Where is your source?

Re: Apple plans to scan US iPhones for child abuse imagery

#363

Post by pseudalopex »

shuckles wrote: post #360
pseudalopex wrote: post #330

Earlier quoted context omitted.

They scan photos in the cloud now. And do you think Apple will announce and let users opt out of client side scanning changes?

As far as is publicly known, only iCloud Mail was being scanned before this announcement. This is consistent with the lack of overall CyberTip reports from Apple historically (orders of magnitudes lower than comparable platforms).

Articles said Apple scanned iCloud photos last year. I see some of them were updated to be less specific.

Re: Apple plans to scan US iPhones for child abuse imagery

#364

Post by Retric »

pseudalopex wrote: post #336
Retric wrote: post #280

Earlier quoted context omitted.

It’s done on Apple’s servers using your iCloud backups. For good or ill this lets people opt out by disabling iCloud. I would hope most privacy conscious people disable iCloud, but that’s another story.

Most of the process is done on customer's devices.[1] [1] https://www.apple.com/child-safety/pdf/CSAM_Detection_Techni...

I was answering this: “Is the comparison done in my phone against a downloaded database, or is a hash of every image on my phone uploaded for comparison on a server?”

To be clear each image, the image’s NeuralHash, and a visual derivative are uploaded to iPhoto. This allows for the inspection of the NeuralHash algorithm used which I actually prefer.

The phone isn’t downloading the hash database.

Re: Apple plans to scan US iPhones for child abuse imagery

#365

Post by leereeves »

vimacs2 wrote: post #351
skinkestek wrote: post #264

Earlier quoted context omitted.

I've taken all vaccines I've been offered including BioNTech/Pfizer but: carefulness regarding new medical procedures is also completely natural. We aren't robots and we need to cut one another some slack. Different persons being mire or less risk averse is even a strength for society.

I don't disagree there, I'm just pointing out that there is a difference between the two cases.

Your argument was wrong, though, because those vaccines aren't being given to children anyway.

Re: Apple plans to scan US iPhones for child abuse imagery

#366

Post by Retric »

headShrinker wrote: post #358
Retric wrote: post #347

Earlier quoted context omitted.

They are using AI techniques as part of perceptual hashing. In other words it’s trying to match files with specific images rather than abstract concepts.

I’m pretty sure you don’t know what you’re talking about. Perceptual matching is not image matching. Perceptual matching is used to sort categories of images. A quick DuckDuckGo will turn up many results. No stretch of the imagination will turn this into a bit for a bit comparison. This is a machine learning algorithm used to categorize images. https://www.ibm.com/blogs/research/2019/10/learning-implicit... Matthew G…

Perceptual hashing has a very specific meaning. It’s for verifying one file is directly related to a different file not an overall classification.

“Perceptual hashing is the use of an algorithm that produces a snippet or fingerprint of various forms of multimedia.[1][2] A perceptual hash is a type of locality-sensitive hash, which is analogous if features of the multimedia are similar.”

https://en.wikipedia.org/wiki/Perceptual_hashing

The goal is to verify a black and white copy of an image is identical to a colored original. Search algorithms want a similar thing so they can validate an image contains a blue car. However, a perceptual hashing algorithm must differentiate between different images containing a blue car while matching a photoshopped copy of the same image.

Re: Apple plans to scan US iPhones for child abuse imagery

#367

Post by headShrinker »

Apple: An Apple NeuralMatch Agent has positively matched images on device "Apple iPhone 12 (Red™)" geotagged at the coup attempt against Prime Minister Trump occurring on January 30th, 2024. Images and documents have been submitted to the Thump Bureau of Investigation for further evaluation. Your Apple account has been permanently suspended.

You can request a review of Apple's actions, however we are receiving higher than normal requests at this time. This means we may be unable to review your account.

Sorry for any inconvenience.

Re: Apple plans to scan US iPhones for child abuse imagery

#368

Post by jeegsy »

partiallypro wrote: post #14

I really don't see how this is going to end well, there could be perfectly innocent photos on someone's phone of their own children doing perfectly normal things that kids do. Like a kid running butt naked around the house, or a photo of something like a rash that is sent to a nurse friend for advice on what it is etc. I'm all for protecting children from being abused, but how are they going to filter what is normal…

We really shouldn't be moving the conversation window into the technicalities just yet without discussing if this is even ok or reasonable in the first place.

Re: Apple plans to scan US iPhones for child abuse imagery

#369

Post by noxer »

Spivak wrote: post #300
LeifCarrotson wrote: post #271

Earlier quoted context omitted.

It would be better if the false positive rate was known to be nonzero. And what, in this context, is "your data?" Is a photo on an iPhone Apple's data? I'd argue that the thing you should do if you get a hit is to delete your data, unless you want to spend enormous amounts of time and money defending yourself from prosecution by people who believe that false positives are vanishingly rare.

I'm talking from the perspective of a service-owner who handles user-generated content. You typically set up PhotoDNA on two parts of your infrastructure. You scan client-side and refuse uploads to prevent the images from getting on your servers in the first place and then you scan on the backend to catch anything that slipped through. You do the first part so you don't have to call the FBI. I really don't care about…

In the real world the system delete the data, deletes the associated user, bans the user and no human will do anything about it. Most likely there isn't even a report made to a human it just goes under ToS violation or something so no one inside the companies aks questions.

No FBI is called. Imagine the financial damage this would do if specialist come over and try to collect evidence and the company is forced to cooperated and work for them, for free ofc. Only the big tech does this. They have whole teams for this and automated system and what not because surely the FBI wont walk into googled data center ever to collect evidence. But all the smaller companies they cant afford to do this and the detection system is actually to protect them form going bankrupt because they accidentally stored some illegal data from a user.

>And to the letter of the law you can't just delete the data if you found CP that one of your customers uploaded without running afoul of mandatory reporting laws.

Yes such laws exist. So they use filter software from third parties. They intentionally offer some generic filter that includes "unwanted content" and doesn't specify what it is when it find something so the companies can wave away the liability as they have no way to know if the image was valuable evidence of a crime or just some LiveLeak gore. They can justify that no human watches it because that's to expensive.

Re: Apple plans to scan US iPhones for child abuse imagery

#370

Post by collaborative »

cm2187 wrote: post #217
shuckles wrote: post #210

Earlier quoted context omitted.

As far as I’m aware, the laws which require cloud service operators to scan for CSAM have been on the books for a long time, and none of the usual orgs have raised a constitutional challenge to them. I’m not a legal expert either, though.

But that's the big difference, cloud providers are scanning the data they hold on their device. Your iphone is not Apple's device. Cloud providers have a liability if they knowingly store illegal content, the data is stored on their property, Apple does not, it is your property. It's like if Ford was going to scan what you store in the trunk of your car.

Oh, the irony. Apple users will finally demand that their phone be "their" device and won't say that they are perfectly happy with Apple making decisions for them because they absolutely trust Apple
Post reply on HN