Apple Collecting Feedback From Dispatchers Receiving False 911 Calls From Skiers

[Wildkraut]

Not saving a life when there are solutions to saving a life are also unacceptable.

Sure thing, but it should be released when it’s reliable and not get in the way of saving other life’s.

The dark figure of delayed emergency call answers because of busy lines with fake iPhone emergency calls is probably much higher. This crap is surely leading to a lot of delayed e.g. stroke or heart attack treatments where every second matters, where every second decides between life, death, wheelchair or bedridden for life.

 

[I7guy]

That is a good point. They can tell Apple how many more false calls they are getting this winter vs last winter. That will give Apple some idea of how severe this issue is (and what kind of liability the company could be facing).

They would have to adjust for the increase in people as normalcy returns as the pandemic fades into the rear view mirror.

As far as the liability, I have a suspicion apple out away a few billion jn the war chest.

 

[I7guy]

Sure thing, but it should be released when it’s reliable and not get in the way of saving other life’s.

The dark figure of delayed emergency call answers because of busy lines with fake iPhone emergency calls is probably much higher. This crap is surely leading to a lot of delayed e.g. stroke or heart attack treatments where every second matters, where every second decides between life, death, wheelchair or bedridden for life.

It’s apple and therefore everything is a mushroom cloud. It’s the trolley car philosophical scenario. While I wish this feature was picture perfect, I’m glad they released it as it has saved lives. Ymmv.

 

[theluggage]

Proof? What about the lives saved? Irrelevant?

Where's the proof of the lives saved?

I mean, real proof - not "crash detection called the paramedics on X occasions => X lives were saved".

It's a lot harder to prove the knock-on effects of swamping the emergency services with false alarms => delays and missed calls => lives not saved, but the resulting dead people are still dead, so maybe the burden of proof should lie with the people introducing the exciting new tech that solves problems people didn't know they have.

Anyway, at least as a passer-by, if I see an accident, I know I don't have to report it because the victim's crash detection will do it for them. Also, no need to tell friends and relatives my itinerary before setting off for a road trip through the wilderness because crash detection - right? What could possibly go wrong?

...Because nothing has unintended consequences.

 

[citysnaps]

No doubt in my mind Apple is paying attention to every single reported crash incident that results in both true and false positive outcomes, collecting the data and circumstances, and modifying their detection and response algorithm over time to ameliorate false positives as much as possible.

Will it ever be 100.0% perfect? Of course not. Expecting 100.0% perfection 100.0% of the time is unreasonable for any system, especially those that involve humans involved in random and highly variable dynamic kinetic situations, where it is simply not possible. Does that mean Apple's crash detection tech should not be used? Of course not.

 

[citysnaps]

Anyway, at least as a passer-by, if I see an accident, I know I don't have to report it because the victim's crash detection will do it for them.

That may be ok for you. But you shouldn't project what your response would be to seeing an accident onto other people, making the assumption everyone else knows about Apple crash detection, and that accident occupants are wearing an Apple Watch with that feature.

 

[Corefile]

It's illegal to dial 9-1-1 without an emergency so are these owners of the false positive being arrested?

 

[I7guy]

Where's the proof of the lives saved?

I mean, real proof - not "crash detection called the paramedics on X occasions => X lives were saved".

[…]

Imo, that’s an absurd statement. All
That has to be proved is that crash detection works as advertised, even with false positives.

Having said that, even airbags don’t go off in every crash.

 

[RyanMTS]

Emergency dispatch centers continue to complain about Apple's new Crash Detection feature triggering an influx of false 911 calls from skiers and snowboarders.

A report today from the New York Post notes that New York's Greene County and Pennsylvania's Carbon County have experienced a burdensome increase in false 911 calls from local ski resorts due to Crash Detection. The feature allows the latest iPhone and Apple Watch models to detect a severe car crash and automatically call emergency services if the user is unresponsive, but it is also activating when some skiers and snowboarders take a tumble.

Given that emergency dispatchers respond to all calls out of an abundance of caution, the influx in false alarms has put a strain on some call centers and could divert personnel and resources away from real emergencies. There have been several reports about the issue in other popular ski resort areas like Colorado, Minnesota, Utah, and British Columbia, Canada since Apple introduced the feature last year.

In response to the report, an Apple spokesperson told the Post that the company was collecting feedback from emergency call centers that have experienced an increase in automated 911 calls due to the feature, but declined to comment further.

Crash Detection is enabled by default on all iPhone 14 models and the latest Apple Watch models, including the Series 8, Ultra, and second-generation SE. When a crash is detected, the iPhone or Apple Watch displays an alert, which users have 10 seconds to act on. If the user is unresponsive, the device begins another 10-second countdown while sounding an alarm and vibrating/tapping, and then calls emergency services. Due to loud surroundings and thick outerwear, however, some users may be unaware that the feature was triggered.

Apple says the feature relies on sensors like the accelerometer and gyroscope in the iPhone and Apple Watch, along with "advanced Apple-designed motion algorithms trained with over a million hours of real-world driving and crash record data" for increased accuracy. As with rollercoasters, the iPhone and Apple Watch may be mistaking the abrupt movement of skiing and snowboarding as a car crash in some situations.

Apple released iOS 16.1.2 in late November with unspecified Crash Detection optimizations for iPhone 14 models, followed by watchOS 9.2 in mid-December with Apple Watch optimizations. It's unclear if these optimizations have led to a reduction in false 911 calls from skiers and snowboarders; in any case, it will likely take some time before all users update their iPhone or Apple Watch to the latest software versions.

Despite this issue, there have already been several reports about the life-saving feature alerting first responders to actual car crashes.

Article Link: Apple Collecting Feedback From Dispatchers Receiving False 911 Calls From Skiers

I can attest that this exact issue happened to me while skiing this week. My watch all of a sudden started going off, and I had to rip off my gloves, and quickly disable it. They need to add geo boundaries to ignore ski resorts.

 

[webkit]

Sure thing, but it should be released when it’s reliable and not get in the way of saving other life’s.

The dark figure of delayed emergency call answers because of busy lines with fake iPhone emergency calls is probably much higher. This crap is surely leading to a lot of delayed e.g. stroke or heart attack treatments where every second matters, where every second decides between life, death, wheelchair or bedridden for life.

This situation needs to be thoroughly investigated, and hopefully all appropriate parties are (or soon will be) doing so. If it can be shown that Apple’s crash detection featuring is overwhelming 911 dispatchers to the point of responses to legitimate 911 calls being delayed, the feature should be immediately recalled/disabled until the "false 911 calls" problem is corrected.

Depending on what is learned, this may turn out to be a feature that was simply not yet ready for prime time. It's one thing to be talking about potential browser or word processing software glitches but this may be putting more lives at risk than it is helping which is a serious issue.

 

[citysnaps]

It's illegal to dial 9-1-1 without an emergency so are these owners of the false positive being arrested?

Would a person be arrested for calling 911 after witnessing a car crash, where the occupants of the car ended up not needing emergency medical attention and could simply walk away after calming down, relaxing and catching their breath for a few minutes?

 

[tYNS]

What a farce this company is becoming. The feature should be instantly shut down. Mostly Everything Tim Cook's apple has done has been a disaster. To the point they need to revert to Jobs designs to return status quo.

 

[I7guy]

Would a person be arrested for calling 911 after witnessing a car crash, where the occupants of the car ended up not needing emergency medical attention and could simply walk away after calming down, relaxing and catching their breath for a few minutes?

The amount of brain power spent on hypothetical irrelevant comments is astounding.

 

[Wildkraut]

That is a good point. They can tell Apple how many more false calls they are getting this winter vs last winter. That will give Apple some idea of how severe this issue is (and what kind of liability the company could be facing).

This is probably not possible because of the current weird situation with corona going up and down, which adds plenty of extra noise to usual emergency calls count.

 

[theluggage]

That may be ok for you. But you shouldn't project what your response would be to seeing an accident onto other people, making the assumption everyone else knows about Apple crash detection, and that accident occupants are wearing an Apple Watch with that feature.

Sure, because Apple only market phones and watches to a vanishingly small proportion of the population, everybody else has a Samsung, who will absolutely never, ever respond with their own version of crash detection, and we shouldn't ever try to anticipate what human beings will do when they act like human beings... Seriously?

All
That has to be proved is that crash detection works as advertised, even with false positives.

"Works as advertised" would imply that it reliably goes off when you are in a crash <i>and</i> doesn't generate false positives. If it can't distinguish skiing from being in a car crash then it doesn't work as advertised.

 

[theluggage]

Would a person be arrested for calling 911 after witnessing a car crash, where the occupants of the car ended up not needing emergency medical attention and could simply walk away after calming down, relaxing and catching their breath for a few minutes?

I don't know about arrested, but I think you'll find the cops would take a dim view of things if they arrived to find the "car crash" was someone skiing down a slalom course miles from the nearest road with no sign of distress.

 

[hybra]

wonder what is happening with somersault divers too

 

[citysnaps]

Sure, because Apple only market phones and watches to a vanishingly small proportion of the population, everybody else has a Samsung, who will absolutely never, ever respond with their own version of crash detection, and we shouldn't ever try to anticipate what human beings will do when they act like human beings... Seriously?

I don't know what you're trying to say. Seriously.

 

[I7guy]

[…]


"Works as advertised" would imply that it reliably goes off when you are in a crash <i>and</i> doesn't generate false positives. If it can't distinguish skiing from being in a car crash then it doesn't work as advertised.

Things aren’t 100%. This is a good feature enough not to throw the baby out with the bath water. And speaking of skiing what happens if you ski into a tree and get knocked unconscious. (Sonny Bono) So yeah there may have to be some false positives.

 

[Danfango]

When it comes to safety, better false positives than false negatives.

 

[SpotOnT]

This is probably not possible because of the current weird situation with corona going up and down, which adds plenty of extra noise to usual emergency calls count.

Why would covid impact the ratio of fake calls from a ski resort?

I see how it would impact the total number of calls, but am less clear on how it would impact the ratio of fake calls

 

[SpotOnT]

When it comes to safety, better false positives than false negatives.

No, because while those 911 rescue centers are wasting their time dealing with the fake call, people who are having a real emergency won’t be able to get the help they need.

It is standard procedure that if someone calls 911 and does not respond, they have to send someone out to investigate. So it isn’t just the 911 center you are holding up but the local police or ski patrol.

And if they are busying looking for that fake call, they won’t be coming to someone who actually does need help.

False positives have real world consequences.

 

[BurgDog]

If it gets a reputation of too many false alarms using up resources, 911 responders may, by policy refuse to respond to them. Sort of what has happened with burglar alarms where they must go through an intermediary first. Boy who cries wolf situation, real alarm gets ignored.

 

[mayga1]

would it be too presumptuous to assume that iPhone Crash Detection was rushed to market without sufficient testing?

But but apple never rushes anything to the market. Those are the samsung guys. Apple perfects it thus, invents it so, this is a perfect product from what they stole from Google pixel.

 

[LD517]

Where can I find a 5 minute roller coaster?!?!

(It’s still not a great situation if this goes off in your pocket, but roller coasters are mostly 60 seconds or less. Also, roller coasters don’t usually crash, so hopefully false alarms are rare.)

Most roller coaster rides are not 1 min or less, wtf are you talking about 🙄