An open letter against Apple's new privacy-invasive client-side content scanning
github.com
github.com
Just be warned, there will be those that unfairly try to cast this as helping the distribution of CP. Expect a headline tomorrow: "Edward Snowden et al Supports Child Porn" - or some other hot take.
A few other things:
1. Vote with your feet - Put your money in the pockets of the people aligned to your values.
2. Actively support projects you align with - If you use some open source hardware/software, try to help out (financially or time).
3. Make others aware - Reach out to those around you and inform them, only then they can make an informed choice themselves.
Not surprising at all.
In Russia, _every_ measure to limit internet freedom was introduced under the pretence of fighting child pornography and extremism.
Then, of course, it was used to block the websites of political opponents.
Also, wasn't the rational for the crypto ban the same? Either terrorism or Child Porn? If you support E2E you're effectively supporting child porn?
https://gigaom.com/2009/06/16/germany-to-vote-on-block-list-...
It is one of few european countries that still have (actively used) anti-blasphemy laws[1] and laws against insulting foreign leaders[2].
[1] https://friendlyatheist.patheos.com/2016/02/27/in-rare-move-...
[2] https://www.theatlantic.com/international/archive/2016/04/ge...
Conversely, there are cases in Germany where calling someone a Nazi would not lead to a fine. One very clear example:
Person A: [Unambiguously asserts a sincere adherence to Nazi ideology]
Person B: Did you hear that, everyone? Person A is a Nazi.
You calling someone Nazi to shut down someone in public actually seems to work well to shut down public discourse judging from what I'm observing in Germany right now. I don't think those defamation lawsuits go anywhere(even though they should).
Attached is a case where someone appealed a judgement at the constitutional court of Germany because he was fined for calling a social worker in a correctional facility a "Trulla"[0]. Think of it as getting fined by the court for contempt by calling a random person a "Karen" in the US. He did say it, but it hardly constitutes an insult to someones honour. The courtcase argued that the way this law is currently used directly contradicts the rights of freedom of speech. Obviously they lost because it would set a precedent rendering this whole charade meaningless.
The way this law works is that it's undefined enough that what constitutes an insult to your honour and can therefore be used whenever someone threatens authority or if authority doesn't like you, you can be fined.
[0] https://www.kostenlose-urteile.de/BVerfG_1-BvR-224919_Erfolg...
We're all here to make ourselves feel good saying we Took A Stand.
In reality, four weeks from now, do you think anybody will still be talking about it?
I made this same mistake. I was pretty convinced that people were taking Copilot seriously, and that there was possibly going to be ramifications for Microsoft. I wasn't particularly looking forward to it, but I rated it as plausible.
One stonewall and a few weeks later, here we are. Hardly a word anymore about it.
I think if there's going to be any real, long-term change, we have to find a way to combat the real threat: life goes on.
(Made me smile to phrase it like that, actually. That wasn't the combat I had in mind...)
It's true, though. Ask anyone in a month whether they take this seriously. They'll say yes. And what's it worth?
It's all we can do, of course. But that's just another way of saying we're powerless.
Like it or not, we aren't in control. And unless we can figure a way to get some measure of control, then what are we doing here, really?
Your three bullet points are wonderful, and I agree entirely with them. And it won't change a damn thing.
I'm not sure what to do about it.
I’ve also been recommending to friends and family to move to droids with microG.
The largest concern I have for them is that they’re stuck between the devil and the deep blue sea.
Most people aren’t going to run a droid without google, or Linux on the desktop.
Me too. I told my wife today that I'll be looking at a feature phone as I'm not sure I can be bothered with jumping through all the hoops required to de-Google an Android phone. I remember a time before mobile phones, I was just fine without one - smartphones aren't that good, just convenient.
This is companies consistently get away with pissing off their customers. It's the same reason politicians get away with pissing off their electorate. Even in a crowded space people seldom get outraged enough to change their buying habits. But in an industry where the options are already limited (Android or iOS) and where people only buy products, at most, once every two years...? Sorry but literally nobody outside your social circle will notice if you stopped buying Apple products.
I get this is a depressing response and I'm not suggesting that you shouldn't follow through with your threat. But I hear this threat posted unironically a lot and it seems to me that people think they have the power to invoke change simply by moaning on a few message boards or buying a competitors products for a couple of years (then that competitor does something to earn our outrage and we're back to the first company again). But real change takes effort. A lot of effort. More effort than most are willing to commit. Hence why the GP rightfully commented that in a months time people will have moved on to their next outrage.
I can imagine at some point a 0day targeting iOS devices that when they connect to a public wifi network with a compromised router, it distributes CP malware/botnet on to their device and then everyone else that connects to that device in the future (muddying the waters).
Similar actions could also be taken against other companies and state entities systems.
And these actions don't need to rely on most of the population to take them.
The more these "kill chain" like systems become automated and cheap to deploy, the more incentive individuals will have to pursue these types of attacks.
That capitulating, "crabs in a barrel" attitude is why things are the way they are. The majority just sits around waiting for something to happen and for some miracle to happen; for a leader to show up, galvanize the masses to change their ways and save the day. At the same time you don't believe that will ever happen, so you stay complacent, non-active and immobile (physically and mentally).
Stop preaching futility and do something positive. Try and be the positive change you want to see. Just because you've given up doesn't mean you have to try and convince somebody else to give up. You're part of the problem.
That's because I honestly can't think of anything better. However that doesn't make my response invalid.
> The majority just sits around waiting for something to happen and for some miracle to happen; for a leader to show up, galvanize the masses to change their ways and save the day. At the same time you don't believe that will ever happen, so you stay complacent, non-active and immobile (physically and mentally).
I think the issue is more that most people are too busy with their own lives to give a crap about a theoretical problem that doesn't visibly and directly affect them. We've head leaders before and it didn't change people's attitudes. If Snowden couldn't influence people's behaviours then what chance does a few random posts on a niche message board have?
> Stop preaching futility and do something positive. Try and be the positive change you want to see. Just because you've given up doesn't mean you have to try and convince somebody else to give up. You're part of the problem.
That's a bit harsh. I'm just as free to post my views are you are to post yours. And if you can't be arsed to get off your seat and campaign for real change then you're in no position to delegate that responsibility onto me.
Also I think you misunderstand the point of my post. I'm not trying to talk people out of action. In fact I've explicitly stated otherwise when I said "It's good that you're principled enough" and "I'm not suggesting that you shouldn't follow through with your threat". I'm just expressing my expectations about the futility of it all. A futile action can still give ourselves comfort that we've done the right thing even if we know it will make no wider change: like how I always vote in each election despite my constituency being so overwhelmingly in favour of the opposition that my vote is literally meaningless.
If yourself switching manufacturer brings yourself comfort that you've done the right thing then that's reason enough to change. But the pragmatic truth of the matter is that it'll take more than that to influence a company as large and successful as Apple.
Yet, I'm not convinced that you can reasonably know this. So if we can all agree that what's happening here with Apple is a bad thing, perhaps it's for the best to refrain from posting pessimistic takes?
All actions have consequences, even posting to a message board. I think it is wise to formulate an intended consequence in mind before you act.
I do get your point, honestly I do, but you can't have it both ways.
My comment was meant more in a "Do you really want to, given that this might have an effect that we both deem negative?" kind of way.
Even Nokia’s website seems confused about it because it claims the 2017 3310 is a 2G (“GSM: 900/1800 MHz“) device. That can’t be true of even a low-end 2017 device. It wouldn’t even work on many modern networks without at least having 3G!
I also refused to get a mobile phone for years, much to the chagrin of my girlfriend(s). It meant I used to come home to an answering machine full of messages and feel much more loved than I ever have with the immediate drip feed of messages I get now.
I wonder, what will I really miss that a book and a dumbphone won't be able to replace in my life? Perhaps the maps.
I’m seriously considering doing that. Make it painful for the local staff by demanding a refund. Don’t take no for an answer for quite some time. Explain that they just broke the product with a remote update that you had no choice in, so thats why you’re only returning it now. When they ask how it broke explain that they started scanning all your content, thereby claiming it as their own. If they want to own the phone, then they need to buy it back. Etc.
Except when you turned on the phone, you agreed to automatic updates. Contract law isn't on your side.
What happened is that I updated based on what Apple claimed the update did, but they were lying. Contract law would seem to be on my side in such a case.
Watching it again I wondered, would they even know what a dial tone is? (or was:)
The biggest reason I dumped Android years ago (the last Android phone I had was a galaxy nexus or nexus one) was because of the lengthy process to reflash a phone with something like aosp or (at the time) cyanogen. Each model was different. Ultimately I decided I didn’t have time for that and battery life was atrocious. Like I had batteries stashed all over the place.
I figure something like CalyxOS seems to have a good feedback and reputation here. And a donation helps them and allows me to bypass hours of tinkering on xd forums.
Ultimately I think I’m locked in iOS for now as my phone is dual sim and the esim is a company provided line. So I will either have to revert to carrying two devices again or just maintain the calyx/pixel as my bailout device in case my job suddenly my vanishes (like the current device I have in safe keeping)
The typical Apple customer is forking over thousands of dollars. They have substantial control. This isn't like a government where you need a plurality of voters to agree before anything starts to happen.
Apple is running software on their phone that will either (a) do nothing or (b) call the police on them. The situation is questionable even for a normal consumer, let alone a nervous one. Maybe most people will weigh up the situation and keep buying Apple. But they are making a choice, they aren't powerless. And some will start asking how important smartphones really are.
I'm having to buy a smartphone to be able to wash my clothes.
It’s not like end users are tripping over themselves to have all their photos scanned. Does Apple just have gobs of money to burn on things like this that will not increase its bottom line by even one penny?
There is probably funded from somewhere, and I’d like to know who is paying for it.
Especially the argument that this can be weaponised really easily worries me, and it looks like they overlooked it.
For example: some joker sends you a Whatsapp with a lewd picture, WhatsApp by default saves all pictures to your photo roll => you are now on the naughty list.
I really hope they come up with a good answer to that one (or just abandon this unholy plan).
I think people used Copilot and realised it was a bit of a nothing burger.
> find a way to combat the real threat: life goes on.
That's the curse of the news cycle. But actions that do live on are monetary ones. When Apple miss sales targets, perhaps they will take note.
> Your three bullet points are wonderful, and I agree entirely with them. And it won't change a damn thing.
I think the last one is the most important, if each of us here can multiply the awareness of the problem, we can start to make real difference. Tech people have a massive influence over what devices ordinary people buy, after all, who will they go to for tech advice?
> I'm not sure what to do about it.
It's a tough problem, but what I do know is that doing nothing will 100% change nothing.
For starters, please give up the defeatist attitude. They deserve the frontlash; even if it serves no purpose, as you describe it.
Apple have placed 'Privacy' at the core of their messaging, in order to sell their high-end products. In comparison to the already excessive pearl clutching based on moral panic within the ecosystem e.g. sanitising and censoring language and apps. This is a scope-creep via thought terminating clichés like 'think of the children' and the boogeymen, which will not be enough of an explanation, when Apple ID's start getting locked out for capturing entirely innocent moments of their children's lives, or the intimate holiday snaps etc. Conversely, someone can weaponise it by sending you content that will besmirch your good name, or send you away for a stint in prison.
Nonetheless, they will be a couple of moves away from either a lawsuit/collective action or a pushback from consumers, especially when the algorithm (wrongly) labels you as paedo, porn baron or worse -- then placing you in a Kafkaesque nightmare to explain yourself.
https://www.apple.com/privacy/docs/A_Day_in_the_Life_of_Your...
Turn off sync to iCloud, make local backups only.
Who is to say politicians who started threatening tech companies publicly haven’t made threats behind closed doors about Apple maybe being on the hook as an accomplice for distribution?
My company only exists because our CEO had input on an executive order years ago. We hardly “made it” in the free market.
The headlines never tell the full story. Media colludes with politics to generate “the right” sentiment. The spec and implementation details aren’t being discussed, just classic speculation on privacy and overreach.
So much for this site having a higher level of discourse and it’s efforts to dissuade repetitive and knee jerk commentary though.
We’d have nicer things if we discussed how things work and instead of what corporate media wants us to discuss.
To me this CP scanning nonsense is Apple waving a carrot in front of the Establishment to get the Anti-Trust heat off, while selling everyone else on further degrading privacy through a slam dunk deliverable to satisfy the short-game players, but that's just me.
Like waves and particles; it starts with individuals changing their behavior.
But we’d all rather sit around debating ephemeral abstraction (a lot like religion), reconfirming math and the physical universe still work as discovered years ago, as we’re all still “rich enough” the bottom falling out hasn’t gotten to us yet, political problems are for the poor.
The below links encapsulate some parts of why Apple cannot be continually trusted with customer data, or viewed as a paragon for privacy/security. There have been numerous examples, past and present e.g. The Fappening, T2 Chip flaws, atrocious leaks of customer data, Pegasus etc.
Some people with technical prowess, although not deluded, buy these devices with limited desire to constantly tweak or circumvent security features, especially when these devices claim to have them baked in and promise not to intrude.
>So much for this site having a higher level of discourse and it’s efforts to dissuade repetitive and knee jerk commentary though.
I am invested in this ecosystem, so the topic has a certain resonance. I resent your implication that my response to the original author was knee-jerk commentary. Furthermore, I responded to their nonchalance with a view to furnish a counterpoint.
https://en.wikipedia.org/wiki/ICloud_leaks_of_celebrity_phot...
https://www.wired.com/story/apple-t2-chip-unfixable-flaw-jai...
https://www.seattletimes.com/business/technology/apple-pays-...
https://www.theguardian.com/technology/2021/jul/21/why-apple...
(This may actually be a Good Thing)
Nope, and it won’t have the slightest impact on Apple’s revenue or earnings either.
Instead of spreading your nihilism, you could be a force for good. We are not powerless. We can make a difference.
Contact your elected officials. Start a movement. Create a direct competitor to Apple. I know it’s tough, but we all just need to come together and stop the negativity.
Excuses or results. You choose.
Sent from my iPhone
The way to stop companies complying with the law is to get the law changed.
Yes you are. Leave the default and don’t enable this for your children’s iPhones, and the iMessage change won’t affect you. Switch off iCloud photo storage and none of your photos will be scanned. You have complete control over this.
This needs to be dealt with at legislative level.
The sad reality right now is that we're actively suggesting that peoples stop buying smart phones, and I think that would mean that any argument will fall on deaf ears. Yes, projects like the PinePhone exists, but they aren't ready for the general public.
As much as I agree with your comment, the sad reality is that I have to pick between Apple and Google. Of those two companies I still trust Apple significantly more.
Pixel phones + GrapheneOS/CalyxOS. Various phones + LineageOS. Various devices + SailfishOS. Librem 5. PinePhone. Fairphone. /e/. F(x)Tec Pro1.
You're greatly underestimating the ability of people to adapt when suggiciently motivated, especially when so many devices can be bought fully set up and most people use only a few apps anyway.
Apple was nice in theory because it provided out-of-the-box privacy and security that laypeople could just go buy.
As much as I dislike this invasion of privacy, I still trust Apple's products, ecosystem & stewardship over the alternatives (for use by whole family).
Although this is the first time in recent memory where not having a viable alternative to consider is irksome as I don't see where the negative push back will come from to prevent further user hostile invasions like this - can only hope it ends here and it's not the start of a slippery slope towards govt surveillance.
This is impossible. Just like modern democratic voting, you don’t get to vote on an individual policy. You vote on a bundle and that bundle almost certainly includes policies (or in this case “features”) that you don’t agree with.
Too bad many people have been trapped within the Apple bubble and are unable to jump ship because they have no idea how the rest works, how they get their data over there and don't have the time to do all this.
This is why Apple IS a monopolist and this is how it ends up being a problem.
I think there is a #4 that could be on the list. People within Apple could try to push back and protest and walk-out or any other means to try to make this fight go viral. However, the media and population write large will push back "... for the children." I've always thought that these problems are more solvable with Whistleblower Awards plus Witness Protection packages for the major enablers. There are lots of people in these rings or with someone who is but they are dependent on those same bad people for their necessities. Also, some of these rings are crime group adjacent and witnesses would need protection.
From what I'm reading, it seems like turning off iCloud (maybe just for photos?) will turn off this scanning. It is unclear to me what server side scanning Apple was/wasn't doing on photos uploaded to iCloud previously/currently. The one thing that occurred to me is that this is almost seems like this is a cya, Section 230 protection in disguise. There has been more discussions about Big Tech and 230, and this is one way to say "Look, we are compliant on our platform. Don't remove our protections or break us up, we are your friend!"
“ Apple's proposed technology works by continuously monitoring photos saved or shared on the user's iPhone, iPad, or Mac. ”
No it doesn’t, photos just saved in the phone are never scanned by either of the systems Apple is implementing, and shared photos are only scanned in two very specific situations.
Also they avoid pointing out the iMessage aspect of this is opt in only, so it only applies to children and only if parents choose to enable it, and it's off by default.
You can completely avoid all scanning entirely by not opting in children for the iMesage scan, which is irrelevant to you if you don’t have children anyway, and not using iCloud Photo Library. Photos you keep on your phone or share using other services will not be scanned. Anyone just reading this letter wouldn't know any of that. How is this promoting informed choice?
Bear in mind the iCloud Photo Library scan is being implemented in order to meet a legal obligation. Apple is required by law to scan photos uploaded to their cloud service.
So really there are two issues here. One is the question of parental authority over optionally enabling the scan for their children’s use of iMessage. That's a legitimate concern over the privacy of children, no question.
The other is whether this is an appropriate way for Apple to comply with US law requiring scanning of uploaded images, and whether that law is appropriate. Again, entirely legitimate concerns that this letter obfuscates completely.
Yes, so as it is today, this system isn't even useful for its stated purpose. Apple are not so stupid that they don't know that. Either it will not always be as it is today, or the stated original purpose isn't the ultimate purpose.
After all, the problem you state has a simple and obvious solution: just scan everything on the device.
The only saving grace is that a full database of copyright material won't fit on a current device, so as long as you avoid offending people in power in your jurisdiction you're probably fine for a few years.
What would happen as some others have pointed out, people will forget about it. Apple will bump out decent Mac products line, along with very good iPhone hardware in a few months time. Which will add points back to their mental model.
"May be Apple is really doing it in good faith"
"They aren't so bad, let's wait and see."
Apple's marketing and PR have changed in the past 5 - 6 years. My guess is their KPI had changed. And they will ( successfully ) white wash any doubt you have. And 2022 Q1 results ( That is the iPhone 13 launch quarter ) will be record breaking again.
And that is not even including competition. I mean for pete sake is Microsoft or Google even trying to compete? ( I give credit to their Surface Laptop and Pixel team, but still far from enough )
Privacy is obviously just a marketing play by Apple but this time, they can't hide it and it might actually hurt their bottom line.
Forget the evil monopolies. Consider GNU/Linux phones, Librem 5 and Pinehone. They may not be ready for an average user, but the HN audience can take advantage of them and help their develoment.
Phones are products for people who use them. What sells phones is not what OS is underneath layers and layers of code, but user experience. If we are to even hope for alternatives, this is where the focus needs to be.
If the OS underneath mattered the Nokia Communicator would have been a runaway success. It wasn’t. It was a useless brick of a computer that almost couldn’t do anything compared to even the first generations of smart phones from Apple and Google.
That's exactly the point,
Maybe you don't know what Plan9 is, but see it like that:
The Phone is just a Terminal. If i want todo business i connect my terminal to my Workplace-servers, every application, datas and settings are there, the calculation heavy stuff and backup is made on the server. If finished i connect to my private plan9-cluster, the Phone is just a bit more than a intelligent Terminal.
The difference is with a plan9 phone you would ~never have anything todo with software ON the phone, let alone having to worry to "sync" to the cloud to make backups, update Apps or need to encrypt the datas in case of loss.
It's a bit like Cloud-gaming or Virtual Desktops with thin-clients, but much much more integrated.
If they could just figure out how to deliver on it ;-)
Install an antenna? You know that before 5G there was a 4,3,2,1,0.5 G?
Most of the attraction of 5G isn't really anything you as a consumer will ever see. Like more advanced spectrum management and more advanced core network components that allow what you can think of as "virtualization with quality of service guarantees" for multiple tenants. On top of that 5G also makes use of higher frequency spectrum which has implications both for use and the design of mobile networks.
To really make life interesting, 5G also aims to enable private 5G RANs. For instance industry in some countries are heavily involved in building their own 5G networks for their own use in geographically limited areas (made possible by allocating higher frequency spectrum for these uses).
The customers that 5G targets isn't so much the consumer as government and industry. For instance by removing the need for building dedicated infrastructure for emergency, defense, law-enforcement, manufacturing etc. which would represent cost savings.
Part of the challenge of 5G is that regulatory authorities don't understand how 5G differs from what they are used to, so many places in the world (with eager lobbying from the telco industry) one will try to keep new players out of the market. (Of course, they might understand, but there may be "motivating factors" to cling to the status quo where spectrum is mostly allocated in large chunks at significant cost and with bureaucracy that ensures smaller players are kept out)
I see the speed difference definitely, that's was the point here remember?
There are all sort of avenues that could be taken for such laws unfortunately. :/
Once we take other big platforms on account, Apple is actually trying to note privacy. Other platforms just scan everything you put in cloud, but Apple tries to limit the information what they acquire, by scanning it on device. And based on their specification, they scan locally only what is going into the cloud. It took me a while to realize, but this is improvement for what has been happening in the industry for years already.
[0] https://www.macobserver.com/analysis/apple-scans-uploaded-co...
Or does that include ancillary data like images in iMessage chats (even if you have iCloud data basically shut off)
I read the scans were done on device but prior to upload. Just unclear what that really means to me.
And I wont be buying the ipad I was thinking about.
Maybe you freedom fighters should have heeded the warnings over the decades of creeping corporate control and stopped buying their products?
Oooh but that dopamine high from new stuff is so addictive!
Now you’re finding yourself in an overturned boat, adrift at sea, no life jacket, and you’re ready to take on monopoly?
The anti-politics bloc incensed their political hands off approach enabled others to wield politics against them.
Good luck.
Given that there are two realistic choices in mobile OS these days, both bound to hardware (and update schedules to a combination of mfg + carrier whim), it’s not a great position to be in as a consumer. And yet, what is the alternative? PinePhone?
Police funding has shot up in America. It's everything else thats been cut.
>what’s the problem.
There's no problem. It's just a drawn out power grab with a weak pretext.
What else in America has been cut? I'd be interested in that very long list (since you're saying that everything else is seeing cuts; the understood meaning being that a very large number of major items are seeing their budgets slashed). The spending data I see year in year out, or across decades, is showing the opposite.
Healthcare spending has skyrocketed over the last 20-30 years, including healthcare spending by the government (eg Medicaid, Medicare). They're not slashing Medicaid or Medicare, those programs have far outrun inflation and are drowning the US budget.
Social Security hasn't been slashed. Even the leading Republicans no longer dare talk seriously about cutting Social Security (30-40 years ago they commonly did). Trump could hardly run away faster from that conversation, it's the third rail of US politics, absolutely nobody dares.
Education spending has not been slashed. US teachers are among the best paid in the world and Americans spend more per capita on education at all levels than just about any other nation (while getting mediocre results for their epic investment, as with healthcare spending).
Defense spending of course continues to rise.
The US welfare state has continued to perpetually expand. US social welfare spending is larger as a share of GDP than it is in Canada; and it's closing in on Britain (the US will catch Britain on social welfare spending as a % of GDP this decade). The US social safety net has gotten larger, not smaller, over the decades. Programs like housing first didn't even exist 30 years ago; food security programs like SNAP continue to get bigger over the decades, they're not vanishing.
US Government spending has continued to soar year by year. Typically 5-8% annual spending increases are normal (just look at the radical spending increases during the Bush and Obama years, or any of the recent budgets). Total government spending (Federal + State + Local) has continued to climb, it has not been slashed or reduced. Total US government spending is taking more out of the US economy than it ever has outside of WW2 - you have to go back to WW2 level spending to find something comparable.
Total US government spending has increased by roughly 225% over the past two decades (more than triple the rate of inflation over that time). The soaring spending shows no sign of letting up.
The major US Government agencies - such as NASA, NSA, FBI, DHS, VA, DoJ, etc - have not had their budgets slashed over the last few decades, they keep climbing year after year.
The only big one I can think of is infrastructure spending, which has not kept up with inflation because both sides have refused to raise gasoline taxes.
What kind of results do we have to show for the massive increase in government spending? Are our streets now lined with gold? Things are better than they have ever been, is that right? Is our quality of life equivalent to Switzerland, Norway, Denmark, Sweden? Because we now have their per capita government spending levels. The government systems of the US are spending the equivalent of 45% of the economy each year.
Aging populations with for profit healthcare will do that.
>US teachers are among the best paid in the world
It's still a comparatively low paying, thankless profession and consistently fails to attract talent because of low pay. Most Engineers, for example, have starting pay similar to a teacher's pay with 15 years of experience. Both jobs require the same level of training and degree (Bachelor's).
>total US government spending is taking more out of the US economy than it ever has,
Government spending is still economic activity. It doesn't 'take money out' of the economy, it contributes to it. It pays workers like any other employer that can then spend that money in the local economies. If your argument is 'government spending is bad' then that's a terrible argument.
In 2018 tech companies reported over 45 million images of CSAM, which was double the amount that was reported the year before.[0] The next year the number of reports went up to 60 million.[1] I wouldn't expect everyone to agree on the proper response to the rapid growth of child-abuse imagery online, but I don't think the problem itself should be dismissed.
[0] https://www.nytimes.com/interactive/2019/09/28/us/child-sex-... or https://web.archive.org/web/https://www.nytimes.com/interact...
[1] https://www.nytimes.com/2020/02/19/podcasts/the-daily/child-...
They are basically Reveune Collectors.
I would like to see all cops under federal jurisdiction. Let the FBI train them.
I know in my county of Marin we have way to many just looking for any slight moving violation.
I have felt for awhile that we also need complete bans in certain kinds of tech.
With the exception of always on cop cameras.
Today, the U.S. collectively spends $100 billion a year on policing and a further $80 billion on incarceration.
Just the spending on policing is larger than what other countries spend on military. Actually, the only country that spends more on military is China [1], after taking in incarceration costs, it is 2.5 times the military expenditure of India.
[1] https://www.sipri.org/publications/2021/sipri-fact-sheets/tr...
"the only country that spends more on military is China"
1 1 United States 778 4.4 -10 3.7 4.8 39
2 2 China [252] 1.9 76 [1.7] [1.7] [13]
3 3 India 72.9 2.1 34 2.9 2.7 3.7
:facepalm:
I assume you missed that. The comparison is not between militaries, but US' police vs the military.
:facepalm:
"National defense spending was 1.267992 trillion yuan, 100% of the budgeted figure. Public security expenses totaled 183.59 billion yuan, 100.2% of the budgeted figure. "
http://www.xinhuanet.com/english/download/20210313budget.pdf
I'm a detective that works exclusively on online child sexual offences.
The short answer to this is "no", although the question doesn't make much sense to me. Policing has always been near the forefront of technology.
Perhaps you could expand more on what "good old fashioned police work" means, in this context?
Through the decades, the respect shown to eyewitness testimony has generally declined, and crimes with no eyewitness evidence are still expected to be solved.
For offences with a huge online aspect there is no prospect of "getting out there" until you work out where "there" is, because it could be anywhere in the world.
For you does the CP protection end justify any means?
Where would you personally draw the line on mass surveillance by LE for the sake of your specific LE goals?
CP aside, are there other crimes that you feel should be folded-in to a dragnet like this?
> "does the CP protection end justify any means?"
Like, is this legitimately a question you think I might answer "yes" to?
This is the equivalent of "do you support the rape of children?".
I'll gladly comment on more specific points if you are genuinely struggling to understand how Apple could honestly implement this system in good faith.
> "does the CP protection end justify any means?"
It's a style of argumentation. Not personal. When trying to find where to draw a line in the sand, one way is to draw a line that almost certainly encompasses us both. We are obliged to consider: if not this line (obviously) then what line?
My intent was to find your limit. Do you have any qualms with what you may do under the law? For you personally, how much erosion of innocents' liberties would be acceptable?
Based on your earlier comment, I was not asking Apple, I was asking a LEO who acts with some but limited justification. Legal and moral. And I would like to ask in good faith about how you see those limits.
>> This is the equivalent of "do you support the rape of children?
No need to make it black and white. Almost nobody supports this. I am sure you don't. Please assume good faith on my part too. There are always trade-offs. How far would you go?
I think it's legitimate for companies to implement automated systems, such as CSAM and spam filtering, to limit the amount of unwanted material on their networks. I don't have any problem with Apple, Google, and Microsoft, checking the hashes of files I upload (or attempt to upload, in Apple's case) to their servers against ICSE. I would have an issue if employees of those companies had unfettered, unaudited access to users files.
Outside of giving my opinion of a specific proposal I don't know what you expect me to say.
Perhaps you could describe your own "limit" to how much avoidable suffering is acceptable to you before you would support automated scanning of uploads. I don't personally believe it's possible to precisely explain an overarching "limit" in situations that balance competing moral and philosophical concerns.
---
I'm being rate limited now due to downvotes, might not be able to respond further
I accept your answer, and acknowledge that different perspectives are valuable.
My own opinion is far less relevant as I probably will not be implementing or executing systems that target information from large swathes of a population for inspection by my organisation.
Nevertheless, let me give it a shot. I asked you about the social cost of false-positives. You reciprocated by asking me about false-negatives. It's tough - on both sides of the equation.
I try to apply weighting. If the dragnet would be targeted to a limited number (say 100) then it could easily be justified, since the relative horror of one over the other is surely in that ball park. Maybe even 1,000.
The problem for me is that mass surveillance such as the subject of this discussion is not numerically constrained. It's trawling the entire ocean floor for the one or two species that may be legally caught.
Before we started trying to push government-sanctioned and unwanted spyware engines on private devices, I imagine the process looked something like that. Is this incorrect?
By the time it gets to the police, there will be an identified crime.
This has been the case for many years. I don't have the numbers to hand but I believe NCMEC receives around the order of 100 million referrals a year.
EDIT: it's 20 million according to https://www.missingkids.org/ourwork/ncmecdata
https://www.missingkids.org/footer/media/keyfacts - around 99% are from tech company referrals, 1% from members of the public
So how can this system guarantee civil freedom? How can it be guaranteed that it won't be exploited by the small number of people in power to actually inspect it and manipulate it?
Google Drive has 1bn users. Google scans content for CSAM already. Shouldn't we be seeing these negative side effects already?
Proponents of these systems can point to thousands upon thousands of actual "wins" (such as identifying teachers, police officers, sports coaches, judges, child minders etc who are pedophiles) and detractors cannot provide actual evidence of their theoretical disadvantages.
No system is perfect, no system "guarantees civil freedom", this is not a fair test. The actual evidence suggests automated scanning for CSAM is a net win for society.
This is the concern. The system is hidden and you have no way to challenge. You are simply “trusting” that this system is working well
Who watches the watchers?
Now, currently this is only applied to iCloud photos which can already be scanned server side, making this seem like a likely step towards end to end encryption of iCloud photos but not a major change from a privacy perspective. What seems like more of a change would be if it extended to non-cloud photos or the iMessage nude detection scanner since those aren’t currently monitored, and in the latter case false positives become a major consideration if it tries to handle new content of this class as well.
This is disingenuous, since up to this point these databases weren't used in systems residing on end user devices, scanning their local files.
The disadvantages aren't merely "theoretical"; they just haven't materialized yet since the future does not yet exist. To ignore these obvious faults by hand-waving them as theoretical is to blatantly ignore valid concerns raised by thousands of informed citizens.
No system is perfect, but that doesn't mean there aren't some systems that simply go too far.
All arguments relating to anything other than this are arguing against something that doesn't exist.
I don't dispute that fictional proposals in the imaginations of HNers might pose a grave (fictional) threat to civil freedoms.
Given that it is just an implementation detail, I guess they can simply change it and then the rest of us will be at peace?
I think it would do you well to see this quote:
> ”Any proposal must be viewed as follows. Do not pay overly much attention to the benefits that might be delivered were the law in question to be properly enforced, rather one needs to consider the harm done by the improper enforcement of this particular piece of legislation, whatever it might be.”
> -Lyndon B. Johnson
Where would these reports come from though? Without these dragnet methods, it would seem like a very simple matter to get away with owning this kind of material.
There is no point in trying to eradicate all such material and bring the number to absolute zero. It's impossible. The only way to semi-ensure this would be absolute slavery of the citizenry, which I hope is a non-goal.
It's a fact that NCMEC referrals have identified teachers who were secretly pedophiles, who were subsequently banned from teaching. Most people see this as a good thing. If you want to do away with this, you have to bring a compelling argument.
I support everybody's right to argue for the type of society they want to see, but there's an arrogance/conspiratorial flavor to a lot of the comments here that suggest they don't really understand what they are up against. There are actual benefits.
I'm a law-and-order guy myself, have applied for (technical) position in police and would even accept a paycut for it.
But it is with this like with other security related things, many people only considers two of the three cornerstones:
- confidentiality
- integrity
- availability
Same here:
Justice is not only about punishing every very bad guy.
It is also very much about not dragging innocent people through the worst accusations an innocent person can be dragged through.
Keeping people safe is not only about rescuing people from hostage situations - it is also about not ramming in the door at someone relaxing on their couch because some random guy or gal called over ip phone.
Too many people already go through huge problems because of false accusations. You say you work in the field so I guess you must have heard about cases were perfectly innocent people commit suicide because their lives got ruined because of what later turned out to be baseless accusations.
Justice is also about not creating a system that creates even more of this suffering, right?
(BTW: I really enjoyed the an0m trick and a few others. I root for you guys, but lets keep it real and not set up the worst footgun for the future since the well meant Jew lists in Europe early last century.)
I don't agree with the characterization that this system will cause baseless accusations.
These systems have been in place for many years and have proven themselves useful and reliable.
Apple have tried to implement it in a way that allows them to turn on E2EE on iCloud, and HN has turned that into a conspiracy theory.
I don't want to do away with anything; I just want "you" out of my devices, where "you" don't belong anyway and where you haven't been up to this point. (Well, you still aren't there because I don't use an iPhone, but hopefully the point is clear.)
Also note that I'm not against the use of such technology in certain situations, such as in unencrypted cloud storage. Though concerns with lack of oversight exist there as well! But installing this on end user devices goes a (huge) step too far.
Your only concern then, is with a future authoritarian policy of your own imagination?
Of all your (mis)leading questions, this one sticks out the most as being of particularly bad faith.
Commenters here are making claims about false positives, arrests of innocents, lived ruined etc. These effects are identical whether uploads are scanned before or after transmission.
What are the different effects of society from scanning the upload before transmission?
> A situation that has the exact same effects on society...?
This is patently false.
You do come off as arguing in bad faith, resorting to debasing any concerns of policy misuse as simply a product of the imagination of the person raising the concern. As if authoritarian policies and erosion of freedom/privacy are something completely baseless, impossible and unheard of, instead of happening all the time.
What is the most obviously different effect on society? I don't know how anyone would even know what CPU calculated their file hash.
Without even discussing policing across racial and poverty lines, one only needs to look at police failing to catch serial killers, rapists, and even just house burglars.
The situation was much worst 10, 20, 30+ years ago. Nowadays police have some of their work done for them by technology and they are at least able to catch a few more criminals. Technology has also helped prevent false arrests in some cases. Apple's content scanning is problematic, for sure, but technology in general has been great for community safety.
Prosecutions based, essentially, on community suspicion are what lead to countless black men being wrongfully convicted of the rape of white women.
Police embracing cutting edge science like DNA sequencing is what allows unreliable antiquated evidence like (gasp) eyewitness testimony to be given it's proper weight.
Perhaps people consider DNA evidence to be "good old-fashioned policing" nowadays but it was within people's lifetimes that it was as new was quantum computing is today.
The sooner the "good old fashioned policing" meme dies the better.
There is no "old fashioned" work. The police is a relatively new concept. And these issues (Child Porn) are really 21 century issues.
> Why does EVERYONE need to be surveilled for the 0.01% (or less?) who don’t behave properly.
Because you think they are surveilling people to catch the bad people. You have too much faith in the system.
So, now NCMEC and any local NCMEC can provide new hashes and identify – possibly even historically – the epicentre of the distribution of a group of images.
Except, if Apple only gets the hashes, what’s to stop a bad actor in a NCMEC from providing non CSAM images to Apple for other purposes?
Seems like this technology should be illegal without a warrant. It’s a wire tap.
I really think people are missing this point. NCMEC's database is not an infallible, audited and trustworthy source of despicable imagery. It's a mess contributed to by thousands of companies, individuals and police.
It's also so intertwined with the FBI that I don't think it's truly correct to call NCMEC independent, given FBI employees work at NCMEC, including on the database.
The FBI's involvement is particularly notable here, given their past usage of mass surveillance.
This is a truly terrifying precedent.
Also, many people trust Apple to do ‘the right thing’. But if Apple are trusting NCMEC and their local equivalents to provide valid CSAM hashes, Apple will never know if their technology is being abused, so they can conveniently point the finger elsewhere if it is.
I can’t imagine this point has escaped their awareness.
Well that's silly of them, thats the motto Google switched to after they gave up on the whole not being evil thing.
( https://www.engadget.com/2015-10-02-alphabet-do-the-right-th... )
:P
How do you know?
All I can say is that I'm not speculating, I know for a fact the database is as broken as I say it is.
As others have stated, do you have proof? Inside knowledge of this supposed reality? Even some major news publication with no direct evidence would be credible enough to support this statement.
Apple uses sophisticated cryptography to make absolutely certain that you cannot hold them accountable for abuses of this system against you, NONE of which are prevented by its complex construction.
The private set intersection is an alternative to sending you a list of bad-image hashes which uses significantly more bandwidth than simply sending you the list. This alternative is superior for Apple because if they distributed the hashes it would be possible for someone to prove that they had begun matching against innocent images (such as ones connected to targeted races or religions, or sought to out particular pseudonyms by targeted images connected to them). It is inferior for the user for precisely the same reasons.
Some might be fooled into thinking the "threshold" behavior, somehow is in their interest: But no, Apple (or parties that have compromised them) can simply register the same images multiple times and bypass it and the privacy (for apple, but not for you) makes it impossible to detect that they've done that.
We are not that naive to believe for a second that this Spyware Engine will do what is claimed.
This is simply viewed as calculated attempt to legalize Spyware Engine on a personal device covered by some bs story intended to get emotional response from the audience to fog the real issue. It is always done this way so no surprises here. All limitations of free speech on the web in Russia were done under umbrella of protecting children and immediately used against political opponents to this very day.
This is discussion about values [1] not technical details! Simply because Spyware may be already installed. It should not become any legal though or morally accepted anyhow.
They talk a lot about complex crypto to protect privacy but the primary thing it's doing is hiding what apple is matching against, which shields them against accountability.
I fully agree that even if the behavior were currently threading the needle it would still be an extremely bad move.
See also this prior post of mine: https://news.ycombinator.com/item?id=28083838
>.. it would still be an extremely bad move.
My view that this should be considered not only as "bad move" but as "criminal activity" Reasons I've described here: https://news.ycombinator.com/item?id=28096948
NCMEC partners are not allowed to share the raw hashes, and I imagine Apple's contract with NCMEC to create a photo-comparison tool that will have auditable code (well, compiled code, but still) includes such a provision to slow or stop CSAM sharing enterprises from completely reverse engineering and cheating the system.
You don't have to have a copy of the database to be fairly confident that your modifications have made a target image non-matching. You would have to have the database in order to gather evidence that the matching was beginning to be used for unlawful, rights violating purposes, such as collecting targets for genocide.
I think it's a safe assumption that this sort of system is only effective against idiots-- which isn't an argument against it: lots of effective anti-crime measures mostly work against idiots. Adding functionality which destroys accountability which at most improves the system against non-targets, however, doesn't seem like an acceptable trade-off.
How can the image hashes take up more space than the images themselves? Are you sure about this?
I have just tested with a fresh profile with a freshly downloaded thunderbird-78.12.0.tar.bz2 (x64 Linux, en-US) using a Burp proxy. Here is the function that does it: https://searchfox.org/mozilla-central/source/toolkit/compone...
Here is the about:config tweak to turn it off: browser.safebrowsing.downloads.remote.url
Here is an example request: POST https://sb-ssl.google.com/safebrowsing/clientreport/download...
Content is Protobuf-encoded payload containing:
- full path to the mailbox including the username and "secret" random profile name (e.g. /home/jenda/.thunderbird/ioi8uk0k.tbtest/Mail/hrach.eu/Inbox)
- SHA-256 of the file
- attachment file name (e.g. 10-test-blahblah.pdf)
(don't get me wrong, "everybody sucks")
Everybody knows that e-mail is not the best medium if privacy is of utmost importance.
In my view this is weaponization of personal devices on a mass scale with clear intent of normalization of surveillance state on a global level.
The fact that this comes after NSO spyware investigation speaks volumes. They don't care about privacy. They care about control over user data.
I hope this will give power to Linux and FOSS movement and be the beginning of serious Apple detox.
The empowerment today is the same professional stance. Explaining the technical facts to my customers. Nobody will like the idea that their phone or personal computer will actively police on the behalf of big brother.
Serious business people don't have to "follow" tech closely. That's why they pay us.
Apple (and all these companies) also operate in a world with a populace that WANTS surveillance for these areas, does not agree that it is a slippery slope, and has voted in governments where that eventually will force Apple to install an even worse back door.
The Linux and FOSS movement have never been up to the end user device challenge, and they too operate in a world with a populace that WANTS surveillance for this stuff.
The solution to true privacy is never going to be “switch to FOSS” for most users, as that’s sort of like saying “get a degree in math”. It will work for some, but it will be an unstable experience.
The solution has to be to a) ensuring democracy survives and b) convincing most people that the right to privacy is more important than finding child pornographers, and c) to vote for representatives that will uphold that right.
Most of them are direct successors of communist apparatus or political project of inner circles with economical power.
Looking at the EU legislators, they will like Apple idea a lot, so my options are limited to technical evangelism for people that care.
If we were to perform a cyber analog of what the founding fathers did for the USA 250 years ago, it would be something along the lines of declaring all software free from the tyranny of corporate control and state oppression. Free in perpetuity so that our digital projections onto hardware shall reside comfortably each in their own pursuit of happiness.
… and then proceed to conduct asymmetric warfare against state and corporate cyber systems alike.
OSS won’t help here. This system must be shut down, and the gov should deal with pedos in a different way.
I mean this does forget the point that absolutely nothing is stopping anyone making a riscv processor which also has such features; we just have to take the manufacturer’s word that they don’t? Riscv isn’t the solution to any of these problems, but hopefully it’ll enable a few more ethical manufacturers to pop up who do make this stuff their core mission.
Probably not, though.
This is the most powerful RISC-V platform you can buy today. It comes as a Mini-ITX board including 16Gb of DDR4 ram, 1Gbps ethernet, usb 3.2, PCI Express and NVMe.
Of course it is less powerful than an x64 machine at the same price point, but it should work reasonably well when paired with an SSD and a graphics card.
Nowadays I just check the benchmarks of a CPU to have a rough idea of it's performance.
> coreboot is a Free Software project aimed at replacing the proprietary BIOS (firmware) found in most computers. coreboot performs a little bit of hardware initialization and then executes additional boot logic, called a payload.
In this case, you or me might be the expert, who can provide solutions for other people such as an "installer" to do whatever.
I personally don't really have a problem with Apple's plan here by the way; I think the way they're doing it is actually quite balanced and reasonable. But other people should be free to make a different personal decision on this. What I would like is to modify some other aspects of my iPhone though, like some things on the keyboard that still bug me after several years.
This is a political issue, IMO. A thousand people protesting outside Apple HQ and/or stores is worth more than 100 thousand consumer striking. IMO, the main non-political avenue here is alternatives, especially FOSS alternatives. It's hard to see a way to widely adopted FOSS phones from here, but if we had a viable alternative things would look different. That's producer action, rather than consumer action.
Make it an issue in elections. Stand outside Apple stores with a sign. Push for a Digital Freedom law. Etc. An Apple labour union, maybe. The conscious consumer game is a dead end. It usually is.
If "vote with your wallet" doesn't work and only causes, say, 1 million people (out of the billion+ customers) to switch off of iOS, doesn't that mean that the rest of the billion+ people really don't care enough? Maybe a few million more accept the tradeoff of getting iOS for 'scanning my photos for csam and having the ability to move off of iOS later if they expand usage or this system otherwise has issues'.
Maybe, just maybe, even if a million people complain about this and apple loses an extreme amount of iPhone sales upwards of 20 million, they can still sell 220 million[0] iPhones to people who don't care. We are certainly the vocal minority here - whether or not that warrants government intervention or otherwise harms our society doesn't really matter when Apple still makes tens of billions in profit a quarter.
0: https://www.macrumors.com/2021/04/07/2021-iphone-sales-forec...
IMO, no. This is silly logic. I don't particularly care that Apple scan my photos.
First, most people don't care about or understand any particular issue well. You might care about the freedom issue here. Another person cares about oceanic biodiversity. Another person cares about factory working conditions. All of them make purchasing decisions that affect all of these. The civilians will always outnumber activists 100-1 on any particular issue.
Second, in this particular case, choices are few and poor. Hence exasperated declarations of going back to dumb phones or DIYing something.
Third people's individual purchasing decisions don't have any effect on the thing they're trying to affect.
In any case, my point was empirical, not theoretical. There are very few significant examples of consumer action amounting to anything. Where it has (eg dolphin safe tuna), consumer action was one piece of a much bigger puzzle, and more about raising awareness than direct. IE people buying dolphin safe tuna also become insistent on regulatory action. By the same logic, you could conclude that people don't care enough about anything. In which case, we come to the same conclusion.
> Apple Inc. issue a statement reaffirming their commitment to end-to-end encryption and to user privacy.
Apple has no commitment to end to end encryption or user privacy; the premise of this letter is incorrect.
https://www.reuters.com/article/us-apple-fbi-icloud-exclusiv...
The data in iCloud is for the most part effectively unencrypted. iCloud Backup serves as a backdoor to e2e messaging, and iCloud Photos aren't e2e at all. If you are using iCloud Photos or iCloud Backup today (on by default and enabled on most iPhones), you are already uploading all of your data to Apple presently in a way that both Apple and the USG can read without a warrant.
This is by design.
The most sure way for someone to obtain child porn images that won't trigger a hit in these known child-abuse-image databases would be for someone to take new photos.
Is there a reason we can be confident that these databases aren't creating a market for novel child abuse at a greater rate than they're taking abusers out of the community? Esp since presumably abusers are only a (small?) subset of people who view these images?
But potentially with the unintended consequence of encouraging the creation of more child abuse material, in order to evade the detection of recognized images.
But this made me wonder: is there an history of people complaining about this sort of things and actually achieving something?
I think this might have happened with MSFT's hailstorm/passport, where in the end industry opposition meant the project was abandoned, but I can't recall other instances.
I'm really hoping this was just a team at Apple that got blinded by all the cool cryptography they came up with to solve a problem and that the higher ups will react if we screech loudly enough.
[1]: https://sneak.berlin/20201112/your-computer-isnt-yours/
How did they react? Isn't it still in effect? Still sending unencrypted messages when apps are opened?
As far as I'm aware, they plan on having OCSP encrypted, and plan on giving users a choice.
I honestly do not get the outrage over the apple's client side scanning. It's outrageous, sure, but... much of the same people seem to have been ok with unencrypted telemetry being transmitted for every executable run. Not to mention that every executable will also be logged at some point, so apple knows every program you've ever run.
The WASP did it to bring the browser vendors to comply to web standards, Netscape ans MS back then. In the EU you could make the case that user group lobbying influenced regulation.
One of the few tiny (and now seemingly completely void) victories I remember was the PIII processor serial number (1999): https://news.ycombinator.com/item?id=10106870
Sure enough, over 72k votes at OneDrive's UserVoice [2] cause MSFT to back down. They still reduce the storage to 5GB, but existing users (myself included) can opt out of it.
[1] https://www.theverge.com/2015/12/11/9890966/microsoft-onedri...
[2] https://onedrive.uservoice.com/forums/262982-onedrive/sugges...
>Apple's proposed technology works by continuously monitoring photos saved or shared on the user's iPhone, iPad, or Mac.
It does a check if it's being uploaded to iCloud Photos. It is not (currently, at least) continuously monitoring photos saved or shared; shared is Messages specific for child accounts.
>Because both checks are performed on the user's device, they have the potential to bypass any end-to-end encryption that would otherwise safeguard the user's privacy.
There is currently no E2E encryption for iCloud, so short of HTTPS... what is this supposed to mean? There is literally no privacy at all if you upload to iCloud currently.
If anything, it feels like Apple's client side system enables something closer to E2EE while maintaining the ability to detect child porn on their platform.
As you say, it’s important to be precise, otherwise people won’t take your arguments seriously. To summarise:
One system involves hash checking. This is performed when photos are being uploaded to iCloud. The hashes are calculated on-device. Private set intersection is used to determine whether the photos being uploaded match anything in the CSAM hash list. Photos that match will be reviewed manually - if I recall correctly, this only occurs after a certain threshold has been reached. This system will only match known CSAM (or whatever else is in the hash list).
The second, independent, system, is a machine learning model which runs on-device to detect sexually explicit photos being received on Messages.app. This system is designed to be run on children’s phones. The system will hide these explicit images by default, and can be configured to alert parents - that feature is designed for under-13s. This system doesn’t use the CSAM hash list.
Should also be noted the second system sends no images to Apple or is reviewed by Apple employees in any way. It's just using the same on device ML that finds dog pics for example.
I think Apple PR screwed up here announcing all these features at once on one page, and not having the foresight they would be conflated. It also didn't help that a security researcher leaked it the night before without many of the details (iCloud only for example).
Context should also be included. CSAM has been happening for years on most (all?) photo hosts, and is probably one of the blockers to full E2E encryption for iCloud photos (I so hope that's coming now!).
Finally, nothing has really changed. Either iOS users trust Apple will use the system only as they have said or they won't. Apple controls the OS, and all the 'what if' scenarios existed before and after this new feature. As described, it is the most privacy preserving implementation of CSAM to date.
This sums up my entire feelings on the matter at this point.
It's become a bit of hysteria due to this.
You are directly contradicting Apple's public statement:
> Apple manually reviews all reports
( https://www.apple.com/child-safety/pdf/CSAM_Detection_Techni... )
Moreover, it was found that in US v. Ackerman that the NCMEC's inspection of a user's private email was an unlawful violation of their fourth amendment rights specifically because AOL passed along the email based on a hit against the NCMEC database without inspecting it. Had AOL inspected it than the repetition of the inspection by NCMEC would simply be a repetition of the search performed by AOL which was permitted under their contract with the customer.
Not only does Apple state that they will "review" the images, they must do so in order to deprive the user of their forth amendment protection against unlawful search by the government.
> and is probably one of the blockers to full E2E encryption for iCloud photos
Apple is free to provide encryption and they are choosing to not do so.
From the GP > The second, independent, system, is a machine learning model which runs on-device to detect sexually explicit photos being received on Messages.app. This system is designed to be run on children’s phones. The system will hide these explicit images by default, and can be configured to alert parents - that feature is designed for under-13s. This system doesn’t use the CSAM hash list.
> Apple is free to provide encryption and they are choosing to not do so.
Sort of. They cited the FBI a few years ago when they put the brakes on adding E2EE to everything. It would also create a huge target for legislation, 'Apple is helping CP people - think of the children!' The new method of CSAM will let Apple add E2EE, and fight off calls for legislation.
I don't think I agree on the "sort of"--
It is unambitious the effective law in the US, established in court, that if the tech company scanning is a product of government coercion than the scanning cannot be performed without a warrant.
In Ackerman Google went out of their way to testify that they were not being coerced by the government, but were scanning out of their own commercial interest (to avoid a public loss of reputation for hosting child porn on their systems).
If Apple wants to defend their failure to act in their user's best interest, they need to do so directly by calling out the governments coercion. Or otherwise, they ought to admit (as google did) that no meaningful government coercion exists and that their actions at the users expense are driven by their commercial interests.
There are two central possibilities: That the searching is coerced, in which case it is an unlawful violation of the user's fourth amendment rights, or that it is not coerced, in which case it is an activity that Apple is freely engaging with for the benefit of their commercial interests.
In either case-- facilitating an unlawful invasion of user's constitutional rights on behalf of the government, or voluntarily invading the privacy of their users for commercial benefit Apple is ethically in the wrong.
I don't doubt that this dichotomy, if widely recognized, would put apple in a difficult position-- but that's a cost of doing business.
This isn't a protection of your privacy. It's because US v. Ackerman (2016) the appeals court ruled that when AOL forwarded an email with an attachment whos hash matched the NCMEC database to law enforcement without anyone looking at it, and law enforcement looked at the email without obtaining a warrant was an unlawful search and had AOL looked at it first (which they can do by virtue of your agreement with them) and gone "yep, thats child porn" and reported it, it wouldn't have been an unlawful search.
Instead, if Apple were to report the existence of a match to law enforcement without divulging the image, which (if the matching is precise enough) should be enough to create probable cause to allow law enforcement to obtain a warrant. The only once a warrant was issued would anyone look at the image. That would be more protective.
What they're actually doing just inserts an unaccoutable additional step that reduced your privacy for the specific purpose of undermining your due process by avoiding the need of law enforcement to evade your privacy.
>It does a check if it's being uploaded to iCloud Photos.
The letter has a perfect wording because it doesn't relay on things unknown or said by the interested side with agenda to push Spyware Engine using some BS cover story. Wording reflects the real danger of such step.
Your continuous attempts go into tiny little details of what Spyware Engine would do at the first stage and support this BS cover story about children is irrelevant to the problem. Besides, How do you know? Did you see the source code?
This cover story has nothing to do with the issue and invented to get emotional response from people and make them stop thinking about what is actually happening.
And what actually happening is an attempt to legalize Spyware Engine that will do completely other things that are claimed and who knows what it will do...
We are not that naive to believe for a second that this Spyware Engine is going to do what it is claimed to do. This attempt of it's legalization is done just to avoid future scandal once it detected.
Again, it doesn't matter what Spyware Engine would do at the moment of installation. What matter is it's existence on a personal device which should be prohibited completely. No one in healthy mind would believe that even if Apple would say that Spyware Engine would not be installed it would indeed not be installed. It should be checked very carefully. It very may be already installed. The fight now is more about legalization of it and values. Not about Apple. Not about what their specific version of Spyware Engine does.
>There is literally no privacy at all if you upload to iCloud currently.
This is exactly the reason why some people do not use iCloud and not going too. The whole idea of putting personal thoughts/ideas/photos into some cloud in open/de-codable format is completely idiotic. So I agree that this malicious practice is done for some time now by Apple . It is unacceptable. It was encouraging people to loose own privacy which is the opposite of what should be in a healthy free democratic society respecting human rights. It was bad long enough but now there is an attempt to make it even worse, because before at least people had a choice to avoid using iCloud and not participate in malicious Apple activity but now it is suggested to leave people without such choice because Spyware Engine is pushed to be installed in their personal devices.
The installation of Spyware Engine will have longing effect on free democratic society respecting human rights. I described such effect here: https://news.ycombinator.com/item?id=28084578
While I like the sentiment of protest, I don't understand this logic of putting it on a closed platform of one of Apple's biggest competitors who also doesn't respect your privacy.
But this is why Microsoft bought it: to get closer to monopoly (GitHub, Copilot, NPM, VS Code, Azure, etc.) making people buy into a platform they feel they cannot escape. I think it is important to tag the parent company when talking about GitHub so users understand who they are really dealing with by hosting their projects on this Git forge and what restrictions this puts on their contributors.
Are Pixel phones the only android phones that allow you to do secure boot with non stock images and basically long term cripple phone functionality once its unlocked?
For me, re-flashing wouldn't even be that much of a hassle since I have contacts and all in my Nextcloud instance and would only have to make a backup of Signal (I barely use my phone - about 2-3 hours total per month), but I just can't get myself to do it.
Should have sticked with my trusted ol' Nokia 3110c I used for a good 13 years, but alas I sent that in to Fairphone for recycling after I bought my Fairphone 3 last year (which I only bought in the first place because finding CNG gas stations is such a hassle here in Germany).
https://www.microsoft.com/en-us/photodna
Which, in turn, is used by FaceBook, Twitter, etc.
Apple just now moved it to scan on the device.
And if you have your iCloud sync toggle on, since your devices are intended to be uploaded, the client is doing the hashing.
IF we can trust that they really scan locally only those files which would end up into the cloud, then this is improvement. But trust is all we have, because the system is already full blackbox.
https://www.apple.com/child-safety/pdf/Expanded_Protections_...
As far as I understand, you can disable this feature, because it is tied to iCloud sync. Based on their spec [1], this feature avoids to do the same as Google and others doing (scan everything on cloud), instead they scan on device, which limits exposed data what Apple sees. So this is improvement compared to other available solutions.
[1] https://www.apple.com/child-safety/pdf/Expanded_Protections_...
The client side/server side also does not matter because iOS users have had to trust Apple implicitly since day 1. All the 'what ifs' existed whether or not Apple added this feature.
I speculate that Apple is going to announce an expansion of E2E to more services at the iPhone event this year, and this feature is getting in front of political complaints that could lead to real privacy destroying legislation/LEO complaints.
there's still our justice system/due-process as well
Imagine if your Tesla reported back to Tesla who was in your car and sent that to the authorities if they thought someone in your car was "illegal".
Also, you can restore an iPhone from what SyncThing can back up. Only an iTunes or equivalent iCloud device backup can. They also can’t back up things like MFA keys / seeds / IVs.
I’m glad ST exists and it’s useful for syncing photos for home use, but it is not a viable iCloud alternative, nor could one be made due to the private iOS API issue.
- not enough people even know about this issue, understand the consequences or remember to act on it the next time a new phone is due - not enough alternatives on the market, and we're probably already boycotting them for other reasons - even with lower sales a company might not see the cause-and-effect
There is a general trend in the industry to take all kinds of liberties with user data. There is pressure on companies to do something about bad things happening in their clouds. Machine learning folks want to apply their tools to big data and the corpus of everyone's photos is probably nice to work with.
So we would have to tackle these issues on a larger scale.
I haven't heard much of complains about Windows sending all your file hashes and most of the files into their servers, what they have been doing a very long time with Windows Defender. It is literally the same what Apple is doing now, but without restrictions.
Google and other cloud service providers have scanned your photos for very long time, by using PhotoDNA. On top of images, Google scans also your emails. People forget and move on.
Google Drive includes backups of your phone, so it is not limited to Google Photos. Literally everything that is stored unencrypted on Google servers.
It nags you once, but you can ask it to stop.
Besides, uploading unseen executable code and scanning photos are far from the same tech. They're very different things.
In this case not buying the product is to avoid a viperous feature and protect house of from being spied on. Any boycott effects are tertiary.
But of course this feature could be abused. But Apple already has all the power it needs to be abusive. They can push whatever software they want to your devices. They can remotely lock and wipe all your data. They already have the power to do all of these things. This statement is simply an announcement that they will be using some of this power to try to stop one of the worst evils in our world. Until they prove that they will abuse this power, I suggest that we let them try.
It is incredibly myopic of Apple to implement this feature. You must consider, as an engineer, how your constructions will be abused and used for bad as well as good.
It operates on iCloud photos. Those are already scanned. If a nation state wanted to flex this before they could have done so.
'No one has done it before, so they probably won't do it in the future now that it we made it easier.'
What the Apple move is doing is showing that they are willing to relax their restraint. It gets tricky because everyone agrees that the specific goal here is honorable, but the manner by which they are using their power to achieve it is generalizable to areas that are less honorable. Once they are willing to use their power to accomplish one highly honorable goal, it's not a big ask for them to use it for a slightly less honorable goal in the future. Iterate that a few times and you can find yourself in a very bad place. It's the classic slippery slope argument - when you know there is a slope that leads to dangerous places, you need to not ever start down it no matter how righteous the motive is in starting down that path in the first place. There's a reason we have the old saying "the path to hell is paved with good intentions".
The existence of power isn't what matters: it's the intention and willingness to exercise it. Apple is now demonstrating that they have changed their stance in how they choose to exercise their extreme power. That's worthy of scrutiny.
For a concrete example of where I expect this to naturally lead to: instead of a database of child pornography being the source of the hashes to search for, the Chinese government provides a set of hashes of all known digital photos of the Tianenmen square protests of 1989. Does it really seem implausible for a government like China's to NOT use this kind of technology for that purpose? It's not hard to cook up similar examples all over the place.
The difference between the "old" content scanning and the new is indeed that they are now willing to "use" the results of that. Facial recognition was client-side only (or so they said), the results of which never left your phone.
Now they're doing content scanning and sending it to themselves as well as others.
In parallel Apple is starting up a growing advertising business, hiring aggressively and expecting that to be a big part of their future revenue. If they're now "allowed" (by its users) to do content scanning _and_ sharing the results, why wouldn't they use those results for themselves to target you with ads?
What Apple is proposing is basically adding a feature to scan any user's phone for a collection of hashes. Even if they say they will only use this for CSAM this sends a strong message to all government agencies around the world that the capability is over there. Maybe for US citizens this doesn't sound dangerous but if I was a minority or a critic of the government on a more authoritarian country I would jump ship from Apple products right away.
But anything that has
the potential to stop
or slow it should be
fairly considered
Shall we start with you?
(And as people do have naked photos of themselves on their phones often[1], make no mistake, the strip search analogy is NOT an exaggeration)
[1] https://en.wikipedia.org/wiki/ICloud_leaks_of_celebrity_phot...
And certainly you've heard of software bugs. You do not think that a small one like "accidentally" "forgetting" to check that a photo had been synced cannot happen, and an "accidental" scan of ALL photos couldn't possibly happen in any situation no matter what bug/corruption/etc? Surely you'll share with the class where YOU hire infallible programmers.
An accidental strip search is a strip search nonetheless.
Luckily, this can all be solved by NOT writing such code. Ever.
Reference: history of literally every dictatorship includes many "reasonable" expansions of power in the name of security/safety/etc...
The slippery slope from Apple checking image hashes to hourly strip-searches seems rather unlikely, which makes your analogy unhelpful.
Completely agree with your second point. All the 'what ifs' have existed forever. Either iOS users trust Apple will only do what stated or they don't. Nothing has changed.
The sub narrow network used for detecting only CP is of course secret, but then we know it can’t be used for revealing pictures of police, activists etc
Opening up the network would mate it even easier to create the adversarials… Im not sure there is a winning position in the approach they used
And easier to keep Apple accountable on not matching anything else than the purpose.
I don’t care if their thing is open or closed source : My iPhone is mine and I don’t want them to do anything on it that can get me arrested because of a false positive.
We must not accept those intrusions.
This thing is as serious as if IKEA regularly sent robots in your house to check all the photo albums on your IKEA shelf to check them against some database. Why does it sounds dumb when it’s IKEA but more acceptable when it’s your phone constructor ?
And if it is flexible, what about false positives? What if I have a pictured of my naked son on my device and I get flagged? Will the picture of my son get uploaded to the cloud "for further analysis" even though I don’t have iCloud enabled and never signed up for this?
Edit: A sibling post has this link, which answers some of these questions: https://www.apple.com/child-safety/pdf/CSAM_Detection_Techni...
I remember the Twitter account takeover hack a year ago, which was possible only because customer support had some 2FA-bypass: https://en.wikipedia.org/wiki/2020_Twitter_account_hijacking
To prevent false-positives, who knows if they have a "review" team that takes a tiny little peek at my naked son. Do you know for sure that no such system is in place or ever will be in place? What if they do get hundreds of thousands of false positives in the future? How would they improve their system if not by reviewing the existing system with real data?
The tech world seems to have a problem reconciling these two views.
From Apple: “The neural network that generates the descriptor is trained through a self-supervised training scheme. Images are perturbed with transformations that keep them perceptually identical to the original, creating an original/perturbed pair.”
With the iOS ecosystem, in my experience having everything backed up to iCloud is the default position, and you have to turn it off manually.
What I don't want is Google performing a scan of the internal memory of my Android device!
Server side they can do whatever they want, and I choose what data they can access, but on my device not.
Generating and sharing bad PR is the only way to get Apple to change something.
Think about the Bad PR around Butterfly Keyboards, MacPro, App Store Policies etc and Apple reversing course on decisions.
>fundamental privacy protections for all users of Apple products.
Users of Apple (should and largely do) accept they are not in control. Apple has always been clear this is a feature not a bug.
This is the logical next step to justify that their lock-in/monopoly has unique features beside making them more money.
Go to https://www.apple.com/child-safety/pdf/CSAM_Detection_Techni...
and then replace
- "National Center for Missing and Exploited Children“ with „the new Trump administration“
- „child pornography“ with „LGBT content“
Doesn't look fun anymore, does it?
(use „the new Clinton administration“ and „images of guns“ if you are conservative)
Op is pointing out what a universal evil Apple's scanning of personal property at scale is.
But I think it def deserves a lot of scrutiny, and I support the effort, so I'm not really complaining.
Even if we have true E2EE, a bad authoritarian regime in the future can simply make it illegal. It's more imperative than ever to prevent such a regime arising in the first place..
If you don’t think it works, check out your local list of Megan’s Law participants.
The important issue here is an attempt of Spyware Engine installation/legalization on personal device.
I think this is much more serious than many people might see it at the moment. I think it is attack on a very fabric of free democratic society where human rights have real meanings. It is done by people who do not share those values or not familiar with them or simply do not care about them or even worse - doing it on purpose.
Many can think of their personal device as their home. Some can even think of it as extension of their mind. It is very personal space and attempts to invade it can/should be considered not less serious then invasion into your private home.
Even critics can admit that your personal device indeed can be in your private home space and connectivity of such personal device should not automatically mean that someone or something like AI under control of someone should access it and spy on you without warrant.
Companies are doing shitty things for some time now and they are getting away with them. This can be a reason why they move forward with unacceptable practices of surveillance. Perhaps those companies do not understand that surveillance is completely incompatible with free democratic societies respecting human rights and there is no way they can succeed in combining two incompatible things unless they wish to repeat China way of doing thing. And I believe the later would not be met gladly I even think not so much possible without using firearms to which people would resist with the same effort like they did many times in history to protect own freedom.
At this stage I think it is much more serious than just writing a letter.
There should be a law protection of your personal computer from any Spyware Engine/Agent for any reason other then warrant to avoid bigger and possibly deadly confrontations. Just like your home as your personal space should be protected from Search without warrant in Fourth Amendment [2]. Your personal device is much more personal then your home in a way and deserve to be guarded accordingly. I am not a lawyer and perhaps Fourth Amendment is already enough but then it should be used properly to prevent Spyware Engines in personal devices.
* Amendment IV
The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no warrants shall issue, but upon probable cause, supported by oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized. * [2]
The Fourth Amendment originally enforced the notion that “each man’s home is his castle”, secure from unreasonable searches and seizures of property by the government. It protects against arbitrary arrests, and is the basis of the law regarding search warrants, stop-and-frisk, safety inspections, wiretaps, and other forms of surveillance, as well as being central to many other criminal law topics and to privacy law. [2]
[1] https://news.ycombinator.com/item?id=28084578
[2] https://www.law.cornell.edu/constitution/fourth_amendment
end of story.
It's like people are being forced to buy an iPhone or something.
This kind of hash checking is done by damn near all cloud providers - Google Photos, Dropbox, Gmail, Discord, Reddit, OneDrive, Facebook, Twitter, you name it. Apple have actually been very reluctant to implement this.
If you don’t like it, you don’t need to enable iCloud Photos. In the exact same way as if you don’t want your images scanned by Dropbox, you don’t upload the photos to Dropbox. It seems reasonable for Apple to implement something like to prevent iCloud becoming a repository for CSAM.
Edit: I’m unable to respond to the comment below this about hashes being performed locally, as I’m rate limited, but here’s my response anyway: The hashes are calculated on the local device, but only for photos which are about to be uploaded to iCloud Photos. The hashes are sent with the photos to iCloud. If a certain number of hashes in someone’s iCloud match (calculated using private set intersection), a manual review will be performed.
When you use a cloud service, you must (should) be aware of the consequence of your files being stored on another computer on the internet.
Until now, your local files on your device were not scanned in an inscrutable manner.
This is a precedent.
There are some valid slippery slope arguments to be made. But, the feature as described and implemented today actually increases privacy for CSAM scanning.
Turn off iCloud photos and it turns off CSAM.