They cheated. They enrolled the first twin and immediately tried the second twin.
I'm curious, not challenging: how is this cheating?
FaceID and TouchID learn as you use it. Apple even states the accuracy of FaceID improves the more you use it. A device is the least secure immediately after enrollment as it hasn’t been used enough times for Apple to do their “fine tuning”.
A proper test would be to have the first twin use the phone for awhile and perform multiple unlocks until it’s working flawlessly. THEN you try to fool it with a twin.
Security should be tested from its weakest point to get a baseline.
Bull. It should be tested as intended.
When I buy a new router (or wireless camera or other connected devices) it comes with a default password. Should I test how easy it is to break into while retaining the default password? Or should I change the password right away?
1) Absolutely! If a wireless router comes with a blank username, a password as 'admin', and unlimited WiFi access post connection to the settings enabled by default it's a poorly conceived design. If access to the wireless settings can be enabled without requiring a new password then it's a poor design. If it only supports WPA despite being a new router then it's a poor design. If the firmware has been comprised and the company hasn't issued a fix this is also a issue since it's a major weak point. This is why the WPA2 bug is potentially a major issue and while you'll see a fix in all of Apple's OS X-based devices in the next update. Hopefully their routers, too, but only one end of the connection needs to have the fix.
2) The egregious fallacy of your example is that you're using a factory set router against an iPhone that has been purposely set up with a security profile that includes both a 6-digit PIN (at minimum) as well as Face ID. You get that, right?
I know Face ID isn't even officially out yet, but I want a Face ID Remote for my Apple TV so that just holding the remote means that apps, menus, Netflix user queues, music playlists, saved locations on videos, and other content switches to my user profile. Is that too much to ask? Probably, but who would've thought that we'd have this tech just a decade after the original iPhone launched?
Very impressive. Windows Hello, which uses RealSense technology from Intel and only ever stated a 1:100,000 in face recognition security (or 1/10th that of Face ID) faired very well with twin tests so I'll only be surprised if they don't do exceptionally well here. I think this tech is one where each increase in statistical accuracy increases its security exponentially.
I’ve suggested every Apple feature being on Apple TV but closed-minded people say it’s useless.
Although that new ring in the remote makes me very curious. I have a feeling the Apple TV team had TouchID planned there but was removed last minute.
The raised white ring around the Menu button on the new Siri Remote has been great for allowing me to more quickly orient the device without accidentally pressing the capacitance touch screen. I no longer need to use a cloth-elastic hair tie over the base of the remote.
Agree about the ring making it easier but those homemade solutions are a little silly. Why don't you guys buy a cover? They solve to many problems.
How do we know that CNN didn't play us/Apple on that test? But being able to differentiate between identical twins using face detection is quite remarkable. Samsung's face unlock being fooled by a picture -- it's got a loooooong way to go. And so do a lot of other Android phones.
Ah but was it fooled by a picture of a twin?
Not sure if you're being sarcastic but 2D images are incompatible with FaceID.
I know Face ID isn't even officially out yet, but I want a Face ID Remote for my Apple TV so that just holding the remote means that apps, menus, Netflix user queues, music playlists, saved locations on videos, and other content switches to my user profile. Is that too much to ask? Probably, but who would've thought that we'd have this tech just a decade after the original iPhone launched?
Very impressive. Windows Hello, which uses RealSense technology from Intel and only ever stated a 1:100,000 in face recognition security (or 1/10th that of Face ID) faired very well with twin tests so I'll only be surprised if they don't do exceptionally well here. I think this tech is one where each increase in statistical accuracy increases its security exponentially.
I’ve suggested every Apple feature being on Apple TV but closed-minded people say it’s useless.
Although that new ring in the remote makes me very curious. I have a feeling the Apple TV team had TouchID planned there but was removed last minute.
The raised white ring around the Menu button on the new Siri Remote has been great for allowing me to more quickly orient the device without accidentally pressing the capacitance touch screen. I no longer need to use a cloth-elastic hair tie over the base of the remote.
Agree about the ring making it easier but those homemade solutions are a little silly. Why don't you guys buy a cover? They solve to many problems.
Because my free solution solved my issue well enough that I didn't need to make a purchase.
They cheated. They enrolled the first twin and immediately tried the second twin.
I'm curious, not challenging: how is this cheating?
Not technically cheating, but certainly taking advantage of what is likely the most vulnerable circumstance for the twin test; that being right after a face is registered, before additional instances of face data are collected to refine the recognition dataset.
Yeah, it's more of a pathological/corner test case and not cheating. No doubt that Apple's testing regimen includes such test cases as part of their white box test scenarios. There was a earlier comment about Face Id potentially becoming less accurate over time. I don't believe this is the case because the neural networks have temporal dependencies that are effectively lease times on the matching algorithms. Matching data, attributes, and associations are refreshed and older data degraded over time. If you started without a beard and then grew a beard the algorithms are not going to continue to use your pre-beard data for matching once you have grown the beard and been authenticated with the beard. It's not a big logical OR/union situation. In the twin sisters case, and with any very identical twins, there is likely a time based and number of samples within a time-boxed sample period that can effectively widen the acceptance window enough to allow either twin to authenticate. I don't know for sure, but Apple's test teams know all about these factors (hence the white box perspective) and certainly performed extensive testing to assess the performance for all of the pathological/corner cases they could envision. That is why the 1:1,000,000 claims by Apple include caveats for such cases.
FWIW you can also toggle off the "Attention Detection" setting for Face ID if you don't want to necessarily actively look at your phone screen to unlock each time, or if you simply want to speed up the process. Note that it does decrease security a bit.
They cheated. They enrolled the first twin and immediately tried the second twin.
I'm curious, not challenging: how is this cheating?
Not technically cheating, but certainly taking advantage of what is likely the most vulnerable circumstance for the twin test; that being right after a face is registered, before additional instances of face data are collected to refine the recognition dataset.
Yeah, it's more of a pathological/corner test case and not cheating. No doubt that Apple's testing regimen includes such test cases as part of their white box test scenarios. There was a earlier comment about Face Id potentially becoming less accurate over time. I don't believe this is the case because the neural networks have temporal dependencies that are effectively lease times on the matching algorithms. Matching data, attributes, and associations are refreshed and older data degraded over time. If you started without a beard and then grew a beard the algorithms are not going to continue to use your pre-beard data for matching once you have grown the beard and been authenticated with the beard. It's not a big logical OR/union situation. In the twin sisters case, and with any very identical twins, there is likely a time based and number of samples within a time-boxed sample period that can effectively widen the acceptance window enough to allow either twin to authenticate. I don't know for sure, but Apple's test teams know all about these factors (hence the white box perspective) and certainly performed extensive testing to assess the performance for all of the pathological/corner cases they could envision. That is why the 1:1,000,000 claims by Apple include caveats for such cases.
I presented that possibility for testing to make sure that cyclical and/or radical changes (examples previous given) wouldn't affect its security. Apple can make the system more or less secure, as they see fit.
Let's say you are an iPhone X user but were in an accident that didn't damage your face but required you to be in a medically induced coma for a few weeks, at which point you lost a lot weight. You'd obviously have to input your passcode at that point, but will the system take the face it sees at that point to start over with the database because it's now a huge change, will it add it to its current database which could potentially allow for less secure authentication if it determines that a wider range is necessary for this user, or will it say something like "Too much time/too much of a change, you will need to input your Apple ID password to authenticate and then re-setup Face ID." How will it work after someone has a stroke that causes one side of their face muscles to relax? While I can't say what the results are, I have to assume that Apple has considered all these rare yet commonplace fringe cases when determining how to design how the SW will react.
It was a worthless test. Why? 1. One twin (twin #1) registers her face on the iPhone X Face ID. 2. The iPhone X is locked by the reporter while standing next to twin #1 (who has her eyes open). 3. Then the reporter moves the iPhone X in front of twin #1 to twin #2. The twins are standing next to each other. 4. Twin #1 had her eyes open and was looking at the phone while it was given to her sister. 5. Who unlocked the iPhone X the second time? * It is impossible to know because twin #1 was looking at the locked iPhone X while standing next to her twin sister. - This CNN test was nonsense. But of course those who want to criticize Apple will be fooled by bad testing like this.
I’ve now seen a whole bunch of first day video reviews. Mostly, Face ID works as promised. One reviewer did find that in bright sunlight, it didn’t always open. Turning away from the sun seems to solve the problem. Other reviewers working outside, didn’t mention it as a problem. All said that it was as fast as the first gen Touch ID, but not quite as fast as the 2nd gen. All were impressed by the speed. None who tried, could fool it with a photo. A couple said that it takes a bit to get used to it, but then, no problem.
i can’t wait for full reviews to come out, though I’m annoyed that anandtech, by far, the best reviewer of computer related stuff, hasn’t yet come out with their iPhone 8/8+ review.
They cheated. They enrolled the first twin and immediately tried the second twin.
I'm curious, not challenging: how is this cheating?
FaceID and TouchID learn as you use it. Apple even states the accuracy of FaceID improves the more you use it. A device is the least secure immediately after enrollment as it hasn’t been used enough times for Apple to do their “fine tuning”.
A proper test would be to have the first twin use the phone for awhile and perform multiple unlocks until it’s working flawlessly. THEN you try to fool it with a twin.
Security should be tested from its weakest point to get a baseline.
Not when it’s ridiculous. How often do you really think that identical twins, even to their earrings, are going to do what this test did? Except for a playful situation as in the video, never. It’s not real. It’s fun Tv though.
I’ve now seen a whole bunch of first day video reviews. Mostly, Face ID works as promised. One reviewer did find that in bright sunlight, it didn’t always open. Turning away from the sun seems to solve the problem. Other reviewers working outside, didn’t mention it as a problem. All said that it was as fast as the first gen Touch ID, but not quite as fast as the 2nd gen. All were impressed by the speed. None who tried, could fool it with a photo. A couple said that it takes a bit to get used to it, but then, no problem.
i can’t wait for full reviews to come out, though I’m annoyed that anandtech, by far, the best reviewer of computer related stuff, hasn’t yet come out with their iPhone 8/8+ review.
The Verge test in sunlight had the iPhone X at belt height next to the belt. That logically would be the edge of the range of the Face ID system. After all, at belt height / next to the belt this is testing the unlock system while looking up at a person's chin. But this is not Chin ID. It is Face ID. Logically, if in bright lighting, the iPhone X should be moved up to eye level so that the system can clearly recognize the face of the user. As far as I know the iPhone X will work at eye level even in bright sunlight.
It was a worthless test. Why? 1. One twin (twin #1) registers her face on the iPhone X Face ID. 2. The iPhone X is locked by the reporter while standing next to twin #1 (who has her eyes open). 3. Then the reporter moves the iPhone X in front of twin #1 to twin #2. The twins are standing next to each other. 4. Twin #1 had her eyes open and was looking at the phone while it was given to her sister. 5. Who unlocked the iPhone X the second time? * It is impossible to know because twin #1 was looking at the locked iPhone X while standing next to her twin sister. - This CNN test was nonsense. But of course those who want to criticize Apple will be fooled by bad testing like this.
Yeah, it's probably not intentional but their methodology was highly flawed, but now it's out there so it'll probably be used by everyone that hates Apple to prove that Face ID sucks.
I’ve now seen a whole bunch of first day video reviews. Mostly, Face ID works as promised. One reviewer did find that in bright sunlight, it didn’t always open. Turning away from the sun seems to solve the problem. Other reviewers working outside, didn’t mention it as a problem. All said that it was as fast as the first gen Touch ID, but not quite as fast as the 2nd gen. All were impressed by the speed. None who tried, could fool it with a photo. A couple said that it takes a bit to get used to it, but then, no problem.
i can’t wait for full reviews to come out, though I’m annoyed that anandtech, by far, the best reviewer of computer related stuff, hasn’t yet come out with their iPhone 8/8+ review.
The sunlight is an interesting scenario. I would've expected that visible light would have a all effect on the results—and maybe it does—so could this be interference with infrared light emitted from the sun being reflected off your face as heat?
Interesting, but we don’t know if those photos are really using strangers. To me, this looks like the commercials where actors are pretending to be “real” people. These could be the same people in both photos, except for the middle one of the heavy woman, where it’s obvious it’s a different person.
I’ve now seen a whole bunch of first day video reviews. Mostly, Face ID works as promised. One reviewer did find that in bright sunlight, it didn’t always open. Turning away from the sun seems to solve the problem. Other reviewers working outside, didn’t mention it as a problem. All said that it was as fast as the first gen Touch ID, but not quite as fast as the 2nd gen. All were impressed by the speed. None who tried, could fool it with a photo. A couple said that it takes a bit to get used to it, but then, no problem.
i can’t wait for full reviews to come out, though I’m annoyed that anandtech, by far, the best reviewer of computer related stuff, hasn’t yet come out with their iPhone 8/8+ review.
The Verge test in sunlight had the iPhone X at belt height next to the belt. That logically would be the edge of the range of the Face ID system. After all, at belt height / next to the belt this is testing the unlock system while looking up at a person's chin. But this is not Chin ID. It is Face ID. Logically, if in bright lighting, the iPhone X should be moved up to eye level so that the system can clearly recognize the face of the user. As far as I know the iPhone X will work at eye level even in bright sunlight.
Yes. That was Niley. I’ve said before that I think the outer edge of this working would be 2.5 feet. Outdoors in direct sunlight, the working range should be shorter.
It was a worthless test. Why? 1. One twin (twin #1) registers her face on the iPhone X Face ID. 2. The iPhone X is locked by the reporter while standing next to twin #1 (who has her eyes open). 3. Then the reporter moves the iPhone X in front of twin #1 to twin #2. The twins are standing next to each other. 4. Twin #1 had her eyes open and was looking at the phone while it was given to her sister. 5. Who unlocked the iPhone X the second time? * It is impossible to know because twin #1 was looking at the locked iPhone X while standing next to her twin sister. - This CNN test was nonsense. But of course those who want to criticize Apple will be fooled by bad testing like this.
Yeah, it's probably not intentional but their methodology was highly flawed, but now it's out there so it'll probably be used by everyone that hates Apple to prove that Face ID sucks.
I’ve now seen a whole bunch of first day video reviews. Mostly, Face ID works as promised. One reviewer did find that in bright sunlight, it didn’t always open. Turning away from the sun seems to solve the problem. Other reviewers working outside, didn’t mention it as a problem. All said that it was as fast as the first gen Touch ID, but not quite as fast as the 2nd gen. All were impressed by the speed. None who tried, could fool it with a photo. A couple said that it takes a bit to get used to it, but then, no problem.
i can’t wait for full reviews to come out, though I’m annoyed that anandtech, by far, the best reviewer of computer related stuff, hasn’t yet come out with their iPhone 8/8+ review.
The sunlight is an interesting scenario. I would've expected that visible light would have a all effect on the results—and maybe it does—so could this be interference with infrared light emitted from the sun being reflected off your face as heat?
The sun puts out a lot of infra red, thankfully (or we wouldn’t be here!). Our faces reflect, as well as absorb infra red. It’s a reason Kodachrome required special filters in the enlarger heads when printing. Otherwise we’d have problems. It would record it on the film, and as prints didn’t, it would cause odd effects, and be very difficult to properly color correct.
a problem is that different skin colors reflect and absorb infra red and other colors differently, particularly ultra violet, also a problem.
While Apple said that’s they tested with different people around the world, comprising different skin colors to get it to work with everyone, you can’t fool physics. I wouldn’t be surprised if everyone does equally well in normal conditions, under intense infra red, black skin would fare worse, as it reflects more back. I don’t know what Apple could do about that without making the blaster much more intense outdoors, if it’s even possible.
a problem is that different skin colors reflect and absorb infra red and other colors differently, particularly ultra violet, also a problem.
Speaking of, remember that this was a problem with the original Apple Watch sensors? I hadn't heard anything about it since it first launched so I wonder if that was improved and/or overblown?
a problem is that different skin colors reflect and absorb infra red and other colors differently, particularly ultra violet, also a problem.
Speaking of, remember that this was a problem with the original Apple Watch sensors? I hadn't heard anything about it since it first launched so I wonder if that was improved and/or overblown?
That’s a good question. I haven’t read anything about it at all. So, maybe that’s a good thing.
Comments
2) The egregious fallacy of your example is that you're using a factory set router against an iPhone that has been purposely set up with a security profile that includes both a 6-digit PIN (at minimum) as well as Face ID. You get that, right?
Not sure if you're being sarcastic but 2D images are incompatible with FaceID.
Yup, Haha
Let's say you are an iPhone X user but were in an accident that didn't damage your face but required you to be in a medically induced coma for a few weeks, at which point you lost a lot weight. You'd obviously have to input your passcode at that point, but will the system take the face it sees at that point to start over with the database because it's now a huge change, will it add it to its current database which could potentially allow for less secure authentication if it determines that a wider range is necessary for this user, or will it say something like "Too much time/too much of a change, you will need to input your Apple ID password to authenticate and then re-setup Face ID." How will it work after someone has a stroke that causes one side of their face muscles to relax? While I can't say what the results are, I have to assume that Apple has considered all these rare yet commonplace fringe cases when determining how to design how the SW will react.
Why?
1. One twin (twin #1) registers her face on the iPhone X Face ID.
2. The iPhone X is locked by the reporter while standing next to twin #1 (who has her eyes open).
3. Then the reporter moves the iPhone X in front of twin #1 to twin #2. The twins are standing next to each other.
4. Twin #1 had her eyes open and was looking at the phone while it was given to her sister.
5. Who unlocked the iPhone X the second time?
* It is impossible to know because twin #1 was looking at the locked iPhone X while standing next to her twin sister.
- This CNN test was nonsense.
But of course those who want to criticize Apple will be fooled by bad testing like this.
i can’t wait for full reviews to come out, though I’m annoyed that anandtech, by far, the best reviewer of computer related stuff, hasn’t yet come out with their iPhone 8/8+ review.
Only right wing nuts. They think Fox news is real.
After all, at belt height / next to the belt this is testing the unlock system while looking up at a person's chin.
But this is not Chin ID. It is Face ID.
Logically, if in bright lighting, the iPhone X should be moved up to eye level so that the system can clearly recognize the face of the user.
As far as I know the iPhone X will work at eye level even in bright sunlight.
The sunlight is an interesting scenario. I would've expected that visible light would have a all effect on the results—and maybe it does—so could this be interference with infrared light emitted from the sun being reflected off your face as heat?
Yes. That was Niley. I’ve said before that I think the outer edge of this working would be 2.5 feet. Outdoors in direct sunlight, the working range should be shorter.
The sun puts out a lot of infra red, thankfully (or we wouldn’t be here!). Our faces reflect, as well as absorb infra red. It’s a reason Kodachrome required special filters in the enlarger heads when printing. Otherwise we’d have problems. It would record it on the film, and as prints didn’t, it would cause odd effects, and be very difficult to properly color correct.
a problem is that different skin colors reflect and absorb infra red and other colors differently, particularly ultra violet, also a problem.
While Apple said that’s they tested with different people around the world, comprising different skin colors to get it to work with everyone, you can’t fool physics. I wouldn’t be surprised if everyone does equally well in normal conditions, under intense infra red, black skin would fare worse, as it reflects more back. I don’t know what Apple could do about that without making the blaster much more intense outdoors, if it’s even possible.
#TwinGate
The internet is on fire with mockery and suddenly everyone has an identical twin and this is a problem.....