4 ms·

Shatters Apple's argument that all of these hurdles are better for security. I wonder if testimony like this could affect any of their antitrust lawsuits or rig

Shatters Apple's argument that all of these hurdles are better for security. I wonder if testimony like this could affect any of their antitrust lawsuits or right to repair lobbying.

ben_w 3y ago
Not "shatters", as while it is a valid counter, it doesn't tell you the relative strengths and weaknesses of the two approaches, only that Apple isn't perfect which should already have been assumed. A stronger counter to Apple's argument is the relative pricing of exploits… but the story I'm remembering is old enough that I don't want to just assume it's still true, even though it's near the top of my search results: https://www.wired.com/story/android-zero-day-more-than-ios-zerodium/ https://www.wired.com/story/android-zero-day-more-than-ios-z...
- flutas 3y ago
  You can see that it's still the case (for Android exploits to be priced higher than Apply). https://zerodium.com/program.html https://zerodium.com/program.html
- calciphus 3y ago
  Pricing in the exploit market is value based, not cost based. You can sell an iOS exploit for more because the people you're targeting with it are generally wealthier.
  - meltyness 3y ago
    I reject this, but the thought makes me smile.
  - ben_w 3y ago
    > You can sell an iOS exploit for more If you could sell it for more, but it seems you can't sell it for more. This implies a large supply of zero-days competing with each other on price.
saagarjha 3y ago
Apple's marketing might have a different stance but engineers on their security team don't really consider these to be security features.
manuelabeledo 3y ago
It reads more like an excuse than the actual reason. Endpoint protection solutions can be installed in iOS devices. The device could also be wiped clean, eliminating the malware. The latter should not be much of an issue in any serious organization. If any executive keeps critical data in a phone, that is already an issue. The former is a hassle, but I have had to use locked down iPhones before, and the tradeoffs are still better than facing an intrusion. The vulnerability and the vector could also have been present in a different form in Android devices. All in all, I don't think this is the response Kaspersky should have come forward with.
- psychphysic 3y ago
  I'm very happy Kaspersky has. And that their released a tool they believe can detect past and present infection with the Triangulation Trojan. I've not idea to what extent it's possible to have a durable trojan on iOS (probably only the makers of such trojans do know). It's absurd to say a company should not blow the whistle on a sophisticated attack when that companys job is just that!
  - manuelabeledo 3y ago
    > It's absurd to say a company should not blow the whistle on a sophisticated attack when that companys job is just that! They should definitely do it. They should also acknowledge that they did a shoddy job. They let the malware run unchecked for several years. It is clear that the safeguards they had in place did not work, not for protection, but especially for detection. Instead, they chose to boost the image of their own products and bash a third party vendor with a questionable reasoning.
- codedokode 3y ago
  > Endpoint protection solutions can be installed in iOS devices. How does "endpoint protection solution" protect from 0-day exploits? I guess it can do that only in marketing materials, not in reality.
  - manuelabeledo 3y ago
criddell 3y ago
It doesn't really shatter anything does it? People here are going to understand that there are trade-offs to every decision made. I suspect iOS is not worse than the more open Android simply because senior management at Kaspersky are using iPhones. If anybody is choosing their platform with security in mind, it has to be them and they are going with iOS.
- sounds 3y ago
  Previous nation-state level hacking on iPhones used exploits available on the (grey?) market. https://blog.google/threat-analysis-group/italian-spyware-vendor-targets-users-in-italy-and-kazakhstan/ https://blog.google/threat-analysis-group/italian-spyware-ve...
  - criddell 3y ago
    And on that same page it says the Android version didn’t even require an exploit. The sneakiest thing that was required on Android was to write the word “Samsung” on the app icon so that users would click it. Near the end, they say: > This campaign is a good reminder that attackers do not always use exploits to achieve the permissions they need.
runlevel1 3y ago
Is Apple making that argument?
throw74775 3y ago
It doesn't shatter anything except Kaspersky's good judgement. If this is true the real question is: Why are Kaspersky's management using iPhones?
- codedokode 3y ago
  Because they were deceived by Apple's quality promises? If Apple really wanted to improve security (instead of just producing marketing claims about it) they would provide anyone with debugging symbols, root privileges and anything else needed for research and debugging.
  - tux1968 3y ago
    > Because they were deceived by Apple's quality promises? The point being, with Kaspersky as security experts, it really does call into question their judgement and expertise.
    - KMag 3y ago
      It's entirely rational to have believed iPhone to be more secure in the past, now believe Android is more secure, and yet remain on iPhone: 1. At some point, weigh probabilities of exploits 2. Update Bayesian priors as new evidence arrives 3. Even if the initial decision currently appears incorrect, there needs to be a high enough difference in probability to justify switching, because in switching, you're still exposed to any persistent exploitation via the old exploits plus new exploits on the new platform Switching back and forth the instant your Bayesian prior swings over/under 50% for Android being more secure than iPhone is a terrible strategy. (Also, you need to risk-weight your various exploit probabilities... security is a multidimensional quantity, so collapsing to a scalar is at least context-/threat-model-dependent.)
      - throw74775 3y ago
        This is irrelevant to the fact that they claim expertise as to why the iPhone is less secure. They aren’t just claiming it’s because of this one exploit or some exploit stats - they are making the claim that it’s because it’s not open source. Since they knew this all along, we can conclude that they have poor judgment.
gabeio 3y ago
> Shatters Apple's argument that all of these hurdles are better for security. Sorry I don't buy that this "shatters" anything besides peoples misguided assumptions that anything can be perfectly secure without being fully disconnected. Apple's iOS 16 supports iphone 8 which was released in 2017, 5 years ago. Apple's iOs 15 supported iphone 6 which was released in 2015, 7 years ago. > Samsung’s previous promise to provide three years of upgrades and ensures millions of Galaxy users have access to the latest features for security, productivity, visual experience and more, for as long as they own their device. https://news.samsung.com/us/samsung-galaxy-os-upgrade-one-ui-android-unpacked-2022/ https://news.samsung.com/us/samsung-galaxy-os-upgrade-one-ui... They only _just_ changed to 4 years, last year. > Samsung will now provide up to five years of security updates to help protect select Galaxy devices They do mention 5 years of updates but only for _select_ galaxy devices (presumably the top of the line). --- I am assuming anyone rooting/flashing is taking way more risks and security concerns into their own hands. But in length of support/security updates alone apple is winning. I also wonder how long it actually takes a vulnerability patch (let's say for a zero day) to get out on android and then through OEM security updates. (I haven't been android in too long to know this.) Apple actually just released a way for them to do this and have already used it once, they call it "Rapid Security Responses" (which you can switch off although idk why you would). https://support.apple.com/en-us/HT204204 https://support.apple.com/en-us/HT204204

Shatters Apple's argument that all of these hurdles are better for security. I wonder if testimony like this could affect any of their antitrust lawsuits or right to repair lobbying.

ben_w 3y ago

Not "shatters", as while it is a valid counter, it doesn't tell you the relative strengths and weaknesses of the two approaches, only that Apple isn't perfect which should already have been assumed. A stronger counter to Apple's argument is the relative pricing of exploits… but the story I'm remembering is old enough that I don't want to just assume it's still true, even though it's near the top of my search results: https://www.wired.com/story/android-zero-day-more-than-ios-zerodium/ https://www.wired.com/story/android-zero-day-more-than-ios-z...

flutas 3y ago
You can see that it's still the case (for Android exploits to be priced higher than Apply). https://zerodium.com/program.html https://zerodium.com/program.html
calciphus 3y ago
Pricing in the exploit market is value based, not cost based. You can sell an iOS exploit for more because the people you're targeting with it are generally wealthier.
- meltyness 3y ago
  I reject this, but the thought makes me smile.
- ben_w 3y ago
  > You can sell an iOS exploit for more If you could sell it for more, but it seems you can't sell it for more. This implies a large supply of zero-days competing with each other on price.

saagarjha 3y ago

Apple's marketing might have a different stance but engineers on their security team don't really consider these to be security features.

manuelabeledo 3y ago

It reads more like an excuse than the actual reason. Endpoint protection solutions can be installed in iOS devices. The device could also be wiped clean, eliminating the malware. The latter should not be much of an issue in any serious organization. If any executive keeps critical data in a phone, that is already an issue. The former is a hassle, but I have had to use locked down iPhones before, and the tradeoffs are still better than facing an intrusion. The vulnerability and the vector could also have been present in a different form in Android devices. All in all, I don't think this is the response Kaspersky should have come forward with.

psychphysic 3y ago
I'm very happy Kaspersky has. And that their released a tool they believe can detect past and present infection with the Triangulation Trojan. I've not idea to what extent it's possible to have a durable trojan on iOS (probably only the makers of such trojans do know). It's absurd to say a company should not blow the whistle on a sophisticated attack when that companys job is just that!
- manuelabeledo 3y ago
  > It's absurd to say a company should not blow the whistle on a sophisticated attack when that companys job is just that! They should definitely do it. They should also acknowledge that they did a shoddy job. They let the malware run unchecked for several years. It is clear that the safeguards they had in place did not work, not for protection, but especially for detection. Instead, they chose to boost the image of their own products and bash a third party vendor with a questionable reasoning.
codedokode 3y ago
> Endpoint protection solutions can be installed in iOS devices. How does "endpoint protection solution" protect from 0-day exploits? I guess it can do that only in marketing materials, not in reality.
- manuelabeledo 3y ago

criddell 3y ago

It doesn't really shatter anything does it? People here are going to understand that there are trade-offs to every decision made. I suspect iOS is not worse than the more open Android simply because senior management at Kaspersky are using iPhones. If anybody is choosing their platform with security in mind, it has to be them and they are going with iOS.

sounds 3y ago
Previous nation-state level hacking on iPhones used exploits available on the (grey?) market. https://blog.google/threat-analysis-group/italian-spyware-vendor-targets-users-in-italy-and-kazakhstan/ https://blog.google/threat-analysis-group/italian-spyware-ve...
- criddell 3y ago
  And on that same page it says the Android version didn’t even require an exploit. The sneakiest thing that was required on Android was to write the word “Samsung” on the app icon so that users would click it. Near the end, they say: > This campaign is a good reminder that attackers do not always use exploits to achieve the permissions they need.

runlevel1 3y ago

Is Apple making that argument?

throw74775 3y ago

It doesn't shatter anything except Kaspersky's good judgement. If this is true the real question is: Why are Kaspersky's management using iPhones?

codedokode 3y ago
Because they were deceived by Apple's quality promises? If Apple really wanted to improve security (instead of just producing marketing claims about it) they would provide anyone with debugging symbols, root privileges and anything else needed for research and debugging.
- tux1968 3y ago
  > Because they were deceived by Apple's quality promises? The point being, with Kaspersky as security experts, it really does call into question their judgement and expertise.
  - KMag 3y ago
    It's entirely rational to have believed iPhone to be more secure in the past, now believe Android is more secure, and yet remain on iPhone: 1. At some point, weigh probabilities of exploits 2. Update Bayesian priors as new evidence arrives 3. Even if the initial decision currently appears incorrect, there needs to be a high enough difference in probability to justify switching, because in switching, you're still exposed to any persistent exploitation via the old exploits plus new exploits on the new platform Switching back and forth the instant your Bayesian prior swings over/under 50% for Android being more secure than iPhone is a terrible strategy. (Also, you need to risk-weight your various exploit probabilities... security is a multidimensional quantity, so collapsing to a scalar is at least context-/threat-model-dependent.)
    - throw74775 3y ago
      This is irrelevant to the fact that they claim expertise as to why the iPhone is less secure. They aren’t just claiming it’s because of this one exploit or some exploit stats - they are making the claim that it’s because it’s not open source. Since they knew this all along, we can conclude that they have poor judgment.

gabeio 3y ago

> Shatters Apple's argument that all of these hurdles are better for security. Sorry I don't buy that this "shatters" anything besides peoples misguided assumptions that anything can be perfectly secure without being fully disconnected. Apple's iOS 16 supports iphone 8 which was released in 2017, 5 years ago. Apple's iOs 15 supported iphone 6 which was released in 2015, 7 years ago. > Samsung’s previous promise to provide three years of upgrades and ensures millions of Galaxy users have access to the latest features for security, productivity, visual experience and more, for as long as they own their device. https://news.samsung.com/us/samsung-galaxy-os-upgrade-one-ui-android-unpacked-2022/ https://news.samsung.com/us/samsung-galaxy-os-upgrade-one-ui... They only _just_ changed to 4 years, last year. > Samsung will now provide up to five years of security updates to help protect select Galaxy devices They do mention 5 years of updates but only for _select_ galaxy devices (presumably the top of the line). --- I am assuming anyone rooting/flashing is taking way more risks and security concerns into their own hands. But in length of support/security updates alone apple is winning. I also wonder how long it actually takes a vulnerability patch (let's say for a zero day) to get out on android and then through OEM security updates. (I haven't been android in too long to know this.) Apple actually just released a way for them to do this and have already used it once, they call it "Rapid Security Responses" (which you can switch off although idk why you would). https://support.apple.com/en-us/HT204204 https://support.apple.com/en-us/HT204204