Hacker News Reader

I've said this in other threads, but I'm probably done buying AirTags, because the anti-stalking features, that I assume you're aware of, are so annoying, that I don't want to add any more to my life.

When I get in my car with my wife, we both get a notification that an AirTag is following us, because my wife is potentially stalking me with her purse and keys. I can silence this notification for only one day. I can't disable it, because that's what a stalkers or abusive boyfriend would do. They randomly beep and boop to notify me that they exist, and I'm being stalked, and to put my phone near the offending AirTag so I can see the serial and last four digits of the owners phone number.

If someone were actually stalking me, I probably wouldn't notice.

I think the whole concept is almost doomed, if you're to be satisfied. The purpose of an AirTag is to track a thing. The desire of a stalker is to track a thing. I don't think these two identical problem spaces can exist together, in some harmonious way, that would satisfy people with your concerns, without constant annoyance. Just imagine if these were more popular and you did something crazy like, got on a bus, or went on a plane, walked any reasonable distance down a street with someone in your proximity.

I would love to hear a solutions that doesn't result in constant notifications. I also think it's completely silly to say "We'll we just have to restrict location track of anything, because there are bad people out there. Period." If the metric is to stop all the bad, then technology needs a huge lobotomy!

Apple just needs to add shared AirTags to iClould family sharing. I think that’d solve your issue and ~90% of other unwanted stalking alerts. They already allow shared “Find My iDevice” and just need to add AirTags to the mix.

I agree, but from previous HN threads, people have said that this would enable people in abusive relationships to be stalked. Also, it appears that some distance is required to trigger the notification, so I imagine stalkers could use the AirTags within this distance. For example, to find which apartment someone lives in, within a building.

I think there will always be headlines about stalking and abuse, and people claiming Apple didn't do enough (even though it's so annoying as is), regardless.

> would enable people in abusive relationships to be stalked

I can see that concern, but family sharing ALREADY lets you share iDevice locations at all time. That's a superior form of stalking compared to AirTags... So I just don't get this concern. It might just be perception though. From a PR perspective this is a bad time to release any feature that could be perceived as reducing AirTag stalking alerts.

Yep. People saying that have no idea.

or just be able to mark an airtag as friendly?

It appears to already be in the mix for me. I see other reports on this thread saying the same.

Family sharing is how you permanently silence one with you. Otherwise the only option is 24h.

> then technology needs a huge lobotomy!

Yes, it does. It really, really does. The absolutely human-toxic nature of most modern consumer tech ("You buy me, I feed you ads, and collect absolutely as much data as I can to improve the ads other people pay me to deliver to you!" model being the core complaint here) has done a huge amount of damage to humanity.

There are upsides, but we culturally seem horrible at actually evaluating things fairly. Always-on devices, constantly sending notifications, are just horrible to every aspect of humans - and the more we learn about just how horrible the are, the better the companies involved refine their ability to "drive their users nuts" for more eyeball-time to view ads, desired or not.

So, yes, if what you get out of my writings are that we should radically pull back on what consumer tech is and can do, you're evaluating my position quite accurately.

> If the metric is to stop all the bad, then technology needs a huge lobotomy!

I think we need to change the metric. Every tool humans have ever invented can be used maliciously. So we have fallback options, like punishment, to mitigate that. Apple needs to make you read and acknowledge a notice when you get an AirTag that says quite clearly that they know the serial number and will happily provide your contact info to authorities on request. And then police need to actually arrest people for misusing tags to stalk.

Why should the public bear the costs of a $2 trillion dollar company's decisions? Why should overpoliced communities be subject to even more policing?

The solution here is to restrict airtags to finding your own stuff only. That takes away some of the utility but also all of the stalking risk.

I think you’re completely misunderstanding the threat model here.

First the reality: Malicious actors interested in, e.g., tracking other people or other people’s things without their knowledge or consent, have many other options. To defeat them, an Apple product just needs to be somewhat less convenient/useful/cost-effective than some of the many other options.

Second, the perception: Apple needs to convince their customers and potential customers that they care and they’ll do something when there’s an issue.

As released, they pretty much had number one covered, though improvements are always good. So they are mostly working on number two… it’s good when there are actual improvements, but that’s not the main point.

Exactly– the SmartThings tracker didn't ship with any of these features. Tile didn't either. The only reason this is such a conversation is because Apple built these anti-stalking features and marketed them.

Not to mention I could probably make a crappy GPS tracker with none of these alerts and stick it to people's stuff. It may be clunky, but there are plenty of good hiding places. Or I could just get this[0]. Or this[1]. Or any of the dozens of trackers I can find on Amazon that would be perfectly good for stalking.

[0]: https://www.amazon.com/Amcrest-LTE-GPS-Tracker-Geo-Fencing/d... [1]: https://www.amazon.com/LandAirSea-Waterproof-Magnetic-Person...

AirTags shipped with anti-abuse features on day one, no other product provides any anti-abuse features whatsoever.

The speaker could be defeated, but the anti tracking stuff cannot be stopped without breaking the basic functionality.

If you’re demanding that companies must go to great length to avoid abuse of their products why aren’t you demanding the same from tile? Why not the myriad gps trackers on Amazon? It’s not unreasonable to demand that Amazon ensures every such tracker they sell has speakers and anti abuse features.

> It’s not unreasonable to demand that Amazon ensures every such tracker they sell has speakers and anti abuse features.

I agree, however, this thread is about Apple's AirTag devices, and complaining about Amazon's incredibly vile sales practices seems at least somewhat off topic for it.

And Apple's device have vastly superior anti-abuse features than every other product, and clearly apple has not stopped fighting said abused.

What's happening now though is people are attacking apple because AirTags anti-abuse features work, and people have identified when they're being stalked.

We haven't heard stories like this about tile, not because it doesn't happen, but because the complete lack of any anti-abuse features mean that victims are unaware they're being tracked.

To what extent is any hardware manufacturer responsible for malicious usage of their product in your opinion? Surely you don't believe that all bases can be covered no matter how simple the product prior to launch. I credit Apple for making a best effort at launch and continuing security and support for much longer than the industry standard in their products. But realistically there is no way to launch something flawless as you seem to be suggesting.

You're both right; it's an interesting question. Something that can't be used maliciously is probably not useful for much of anything at all. OTOH, when the potential for harm is far greater than the potential for good, as it arguably is here, that calls for extra discretion on the part of the developers.

> as it arguably is here

That may be understating it a bit. I honestly can't decide which is greater.

Airtags as-is are nerfed for non-technical reasons as to make them more or less useless to me, despite their ability to be incredibly useful.

For example:

Due to the nag/anti-stalking alerts I can't put them on dog collars due to day-care visits. No way for others to whitelist a tag on their phone, or any way to silence it.

Same goes now for the car. It's annoying to just toss one in each trunk in case of theft or whatnot, since the cars are shared.

The privacy concerns have more or less turned this product into something of a very narrow usage band - aside from a personal bookbag or whatnot I don't see many other realistic uses at this time. Not useful for hiding in my power tools, etc. etc.

I had pretty high hopes for these for some peace of mind (dogs), and I live in a high crime area so being able to track my expensive items has proven useful in the past. I was excited about expanding on this use until they started crippling them.

Meanwhile, you can get a cheap GPS tracker for the price of an AirTag or two…

And if you want it to report out, it has to have an active cell plan, and the power consumption is rather radically higher than "years on a button cell," so the device is bigger, it needs a clear enough view of the sky to get a GPS signal, etc. They're an awful lot less useful and less discreet. Again, not impossible by any means, but certainly quite a bit less convenient than an AirTag with a removed speaker.

Or just get a Tile, that will ping any Amazon echo or Ring doorbell it goes near.

I just checked and all the reports of misuse so far don’t seem to be “for months at a time”. I can also get a prepaid sim with very little identification and there’s no direct link from an arbitrary gps device to the person who used it.

I think there was a brief surge in people abusing them because they thought AirTags were untraceable, which they very explicitly are not. As apple says in the article they can directly go from an AirTag to the account that owns it.

> I just checked and all the reports of misuse so far don’t seem to be “for months at a time”.

Its also only like 6-8 months old at this point.

I don't think the typical issue is that someone can maliciously track some other person specifically for years. Longevity of the tracker is an added benefit, not what makes it different.

Honestly, in my personal opinion I feel that only thing that's really different is the company's notoriety. It's no fun to bash Tile (people just don't have any beef with them), but kinda fun when it's Apple. And that they've actually sent out a viral idea ahead of product release focusing attention on this aspect (so, naturally, it echoes back).

That was kind of my point. People can buy these GPS trackers for the same price and have a stealthy surreptitious tracker. What’s new is Apple made so omg, better freak out and write eleventy-one angsty articles about them. In this case I can’t even muster an eyeroll at how silly it is.

If I were actually interested in tracking someone for nefarious reasons, I absolutely would not use an AirTag. I mean, it has a serial number that will lead the police right back to me. I'd use something that doesn't have my name on it.

I think that a key issue in this case isn't just that Apple made a device that could be used for tracking - after all, you could engineer or buy a different tracking device from a number of vendors - but that its feasibility was massively increased due to a unique Apple asset - its network of iDevices. That is, Apple released a product that was uniquely good at tracking people due to an Apple-specific asset.

Slightly less, but still very importantly, there's the fact that by design you can't see if someone is tracking you unless you buy another Apple product, which wasn't an intrinsic engineering limitation, but just a choice that Apple made for their own profit.

Apple released an Android app that lets you check for AirTags separated from their owner near you: https://play.google.com/store/apps/details?id=com.apple.trac...

Of course you have to know about it and it only provides scanning when the app is open instead of constantly doing it in the background like iOS.

Makes me wonder how much liability the owner or participants in the network used for tracking carry if bad things are facilitated by the network.

> its feasibility was massively increased

As was the risk of getting jail time if you use it to stalk someone. Something you don't really worry about with a generic GPS tracker. That handy, convenient AirTag will give the police your address.

I don't think all bases can be covered, certainly. But neither do I think Apple has been doing a good job of even trying to do a good job of it, at least recently.

Apple clearly put some thought into how the product would be misused, and added some features for that - but then appears to have not bothered having actual (simulated malicious) users test those features to see how they'd bypass them. Things like "removing the speaker" seem oddly trivial, yet there's no indication Apple even thought through that situation. It wouldn't be too difficult to design something in which the deliberate disabling of the speaker rendered the device useless after a few days, yet would be unlikely to trip in normal use. They're pretty well sealed.

"Red teaming" something like this in the early design phases is often useful to be able to figure out how to mitigate these sorts of attacks, and Apple, far too often lately, seems to be in a "...they did what? No, they can't have... we didn't think they'd... ugh, OK, let's add something to support that..." mode.

And if they don't have family sharing support, as seems to be the case from the other comments here, it clearly means they weren't tested with any sort of realistic use case, because "Oh, yeah, my wife's keys in her purse keep making my phone go off when we're driving together" seems a common use case to discover in testing.

I don't think a company should be responsible for all malicious uses, but when those uses seem utterly trivial to manage (remove speaker, your tracking target uses Android and like almost all Android users hasn't downloaded Apple's app to check for AirTags following them), I think they've missed something really important in the design phase.

Making AirTags safe is fundamentally impossible, a bit like trying to make a knife impossible to hurt yourself with.

For example:

> deliberate disabling of the speaker rendered the device useless after a few days

That only kicks the can. Now people will silence the speaker from outside, for example by covering the whole AirTags in superglue and foam to muffle all vibrations.

> for example by covering the whole AirTags in superglue and foam to muffle all vibrations.

Which makes it larger, harder to hide, and easier to find. Go try - see how much foam it takes to reasonably muffle the beeper. It'll take a lot more than you think.

If I put it in your bag or in your car, I can make it 10 times bigger and you will still not notice it.

Use your own thinking, how would a red team go about silencing the device.

Also, try covering your phone speaker holes with your finger. See how effective that is.

I'd really want to ask weird and possibly insensitive, but nonetheless (I believe) a legit question - shouldn't be it a society that needs to somehow change (I've no idea how, though), rather than technologies killed because it threats existing societal stuff? I hate to say this but it really feels that our assumptions of privacy are becoming outdated and there's no going back.

Unlike e.g. facial recognition, AirTags are not even some new breakthrough, just some well-known existing technologies made a bit more mainstream/commonly available (can't even say "more affordable", the price point hadn't changed). Yet, it causes an uproar.

Technological genie just cannot be put back in the bottle and it's only gonna get worse in the future, as tech will become smaller, smarter and more magical.

> ...shouldn't be it a society that needs to somehow change (I've no idea how, though), rather than technologies killed because it threats existing societal stuff?

We've allowed, collectively, anything that falls in the "tech" realm to deploy first, ask questions later, and more or less force society to conform. I think it's been absurd, we see more and more backlash against it, and rightly so. A company shouldn't be able to just ignore the law and do whatever they want because they use computer chips, but we've seen more or less exactly that in a number of cases.

- Uber/AirBnB can be seen as venture-capital funded, money losing ways around existing taxi/hotel/rental laws, and their tendency to expand into a market without permission and then demand the right to stay there against the laws that prohibit exactly such things, is an example of this model. - The various "Dump tons of scooters in a city without asking, profit, and make cleanup the city's problem" is another example here. It's littering, really, but somehow tech companies involved get a pass because it's neat. - Literally every "We collect all the behavioral data you don't realize we're collecting to improve our models of you so we can sell better advertisements to our actual customers" business model does the same thing - take first, ask permission never (or, at least, try very hard to never have the topic come up).

I'm not OK with this, and I hope more and more people aren't OK with it either.

> I hate to say this but it really feels that our assumptions of privacy are becoming outdated and there's no going back.

Privacy shouldn't be something subject to the current state of technological surveillance. We should, as a society, decide what privacy ought to be, and then limit tech companies accordingly.

Europe seems to be making decent headway on this front, with "No, you can't just keep doing that..." rulings against tech companies. I hope that spreads.

> I'm not OK with this, and I hope more and more people aren't OK with it either.

Cool, that's a societal change - awareness.

I can't argue against your ridesharing datamining examples. I can't say the data they've hoarded is actually valuable - just based on the fact almost all their ads are complete misses. I've yet to see any personal data hoarder company to be able to actually pull off anything sensible, not that some ML model realizing nonsense in lines of "meatbag bought a sofa, meatbag must love sofas, let me show him all the sofa recommendations" (lol). Anyway, that's not the point. I don't fancy this data collection and sure agree it's not a bad thing to stop it. Just as I'm fed up with small prints, shrinkwrap licenses and other things we pretend people read and understand that they don't.

When I wrote my comment I thought about somewhat different things. Tracker devices or face recognition is different because it can be done in what marketing calls a "privacy-respecting" way, on device. Yet, it's a fact if someone can point a camera on you, they can match you with their previous data, and if someone can plant a device on you (and devices get smaller and smaller every few years) they can know your location. And it's not about companies producing something, it's about the technologies itself, when the technology can be used for both good and bad things. Like a knife or a laser pointer.

It doesn't make sense to me to tell companies to not produce such technologies. It's impossible to make technologies be usable for good things and not usable for bad things. A knife is always a weapon (or it's not a knife and you can't use it to make yourself a sandwich), a key tracker can always be used to stalk (or it won't be ever useful to find stolen things). But some people are so desperate for "safety" they do want to nerf anything that can be even theoretically abused to absolute uselessness.

Things like this is what bothers me.

There are lots of tracking solutions out there, most of which do not even attempt to protect against abuse. So, Apple seems one step ahead of the competition here.

Regarding CSAM, I have not seen any credible threat model with those hash collisions.

> "Buy an iPhone to prevent people from tracking you, and if you want to see where the tracker is, buy a really new one!"

The tracker beeps, and displays information to Android phones, also. They did what they could to make it easily findable.

Have any of the AirTag stalking mitigations actually been defeated by hackers? Besides physically removing the speaker.

The “an AirTag is traveling with you” alerts haven’t been defeated as far as I can tell. Apple tweaked some of the time periods after launch but the core technical mitigations are the same.

I agree the current situation is non-ideal, but it’s not clear to me this is from a lack of sufficient pre-release pentesting.

> The “an AirTag is traveling with you” alerts haven’t been defeated as far as I can tell.

Perhaps not, but they're meaningless to the >50% of the population that doesn't have a modern enough iPhone as their daily carry phone.

Having moved back to a flip phone, I literally have no way to identify if such a tag is being used against me.

And 100% of the population has no way to monitor if a GPS tracker has been planted.

GPS trackers are a lot pricier and often bulkier though, somewhat limiting the probability of it to be used by the average stalker. Barrier to entry is significantly higher.

> GPS trackers are a lot pricier and often bulkier though

A quick look on Amazon UK suggests that there are plenty of cheaper options and, whilst not many, there are options of a similar size to the AirTags.

Now, admittedly, none of them have the same feature set as an AirTag but if you're at the point of buying GPS trackers to stalk someone, I doubt you're thinking about the technical distinctions.