This is not like the Black Mirror episode, Nosedive (or a similar episode of the Orville), because people aren’t rating others and scores aren’t known at all. The behavior changes from constantly rating and being rated by lots of subjective raters is what scared me in the episode.
This is similar to other risk or trust algorithm used by banks, healthcare, ctrfit agencies, email services, etc.
It could be bad depending on what Apple does with it, obviously, but this seems like a way integrate more device data into their fraud detection methods.
There's an obvious problem with enabling people to opt-out of fraud detection. But I suppose Apple could offer other ways to confirm your identity, e.g. 2FA.
How can I not use either Android or iOS? I'm not well versed in data protection policies, but at least in the EU where I live, shouldn't this fall under GDPR? I should be able to opt out of having my personal data to be used by a algorithm to determine my "trust score", whatever that means.
The quote from Apple is that they're computing trust scores for your devices for when you attempt a purchase. It sounds like the more you use the device, the more trusted it is. That brand new device that you've never used before shouldn't be trusted.
What I don't understand is how you can reasonably stretch that out to giving people trust scores.
"To help identify and prevent fraud, information about how you use your device, including the approximate number of phone calls or emails you send and receive, will be used to compute a device trust score when you attempt a purchase," the page reads.
See the "information about how you use your device, including the approximate number of phone calls or emails you send and receive".
So the score calculation for that device includes the approximate number of phone calls or emails you send and receive on it.
You could interpret that as "the more personal activity you have on a device, the less trusted it should be", but that makes no sense. The more you use a device, the more likely it is to actually be your device.
It makes sense to place less trust in devices that you rarely use.
Phone calls and email? So only senior citizens will be trusted? What about texts? Are calls and apps measured using only Apple apps? What if I used 3rd party apps for these functions?
I get very few phone calls, and half of them are spam calls that I don't answer (only answer calls from known numbers). I don't use Apple Mail. So I'm untrustworthy because my social configuration isn't typical?
My ipad isn't used for calls, messaging or email. Does that mean I can't use my ipad for purchases?
Have to have ID and credit history disclosed to get a phone in the first place. Isn't that enough? Is this only for non-phone devices?
This needs more in-depth reporting. What kind of fraud exactly?
How many phone calls / email does one send to be deemed trustworthy?
And on top of that, considering I make ‘phone calls’ via signal, FaceTime (audio) and other apps, on top of the standard dialer, and for email use Zoho, inbox, gmail and the standard client, how does it even come up with these numbers?
I have to assume they’re looking for extremely high usage, robo dialers or spam outfits using iPhones. But is there any evidence this is happening?
It also sounds like it’s somehow connected to Apple Pay, but why would someone using their device for spam purposes have or setup Apple Pay? Surely if you’re using your device to robo dial / spam, it’s one of many devices sitting in a rack / pseudo data centre. Not a personal device.
15 comments
[ 1.2 ms ] story [ 58.3 ms ] threadThis is similar to other risk or trust algorithm used by banks, healthcare, ctrfit agencies, email services, etc.
It could be bad depending on what Apple does with it, obviously, but this seems like a way integrate more device data into their fraud detection methods.
The quote from Apple is that they're computing trust scores for your devices for when you attempt a purchase. It sounds like the more you use the device, the more trusted it is. That brand new device that you've never used before shouldn't be trusted.
What I don't understand is how you can reasonably stretch that out to giving people trust scores.
Click bait?
Where did you see it? It says nothing about how the score is calculated.
See the "information about how you use your device, including the approximate number of phone calls or emails you send and receive".
So the score calculation for that device includes the approximate number of phone calls or emails you send and receive on it.
You could interpret that as "the more personal activity you have on a device, the less trusted it should be", but that makes no sense. The more you use a device, the more likely it is to actually be your device.
It makes sense to place less trust in devices that you rarely use.
I get very few phone calls, and half of them are spam calls that I don't answer (only answer calls from known numbers). I don't use Apple Mail. So I'm untrustworthy because my social configuration isn't typical?
My ipad isn't used for calls, messaging or email. Does that mean I can't use my ipad for purchases?
Have to have ID and credit history disclosed to get a phone in the first place. Isn't that enough? Is this only for non-phone devices?
This needs more in-depth reporting. What kind of fraud exactly?
And on top of that, considering I make ‘phone calls’ via signal, FaceTime (audio) and other apps, on top of the standard dialer, and for email use Zoho, inbox, gmail and the standard client, how does it even come up with these numbers?
I have to assume they’re looking for extremely high usage, robo dialers or spam outfits using iPhones. But is there any evidence this is happening?
It also sounds like it’s somehow connected to Apple Pay, but why would someone using their device for spam purposes have or setup Apple Pay? Surely if you’re using your device to robo dial / spam, it’s one of many devices sitting in a rack / pseudo data centre. Not a personal device.
This all sounds very very strange.