4 ms·

A software control prevents your iphone from spying on you, and a software control prevents your echo from spying on you. In both cases you aren't really sure

A software control prevents your iphone from spying on you, and a software control prevents your echo from spying on you. In both cases you aren't really sure it's the case, you wouldn't know if it were suddenly disregarded or circumvented, and you're just trusting in the good behavior of your chosen provider.

_jal 8y ago
You also don't know that you're not being tailed by a PI, that your neighbors are not focusing a mic on your home, or that your dentist isn't on the lizard-peoples' payroll. However, we all make certain risk tradeoffs based on numerous things, including proxy signals. For instance, I would be intensely surprised to find out my downstairs neighbor was recoding me, based on multiple years of interactions. Likewise, I find the signal Apple sends ("We need a US-based GDPR") about privacy rather more compelling than the signal Amazon sends ("scour the mic for hints about product we can shift"). I mean, sure, while both $huge_bank and your hypothetical deadbeat relative might both steal your money, I bet I know which one you choose to to leave your money with.
- menacingly 8y ago
  I'm just saying it boils down to trust either way, not weighing in on which way that trust should lean.
- 52-6F-62 8y ago
  > You also don't know that you're not being tailed by a PI, that your neighbours are not focusing a mic on your home, or that your dentist isn't on the lizard-peoples' payroll. Totally tangential, so feel free to disregard: This statement reminded me of my experience during my brother's TS clearance (Canada). CSIS readily had accurate data (in surprising detail) about my time spent during a short period they must have observed his family members to use during an interview with him. It was pretty fascinating, mildly disconcerting— but in this case didn't bother me (in the context, I understood why. I certainly wasn't interesting at the time). They only could have captured the data they had by a combination of tailing (however brief or otherwise more expensive much more technical means) and bank records. I was of course none the wiser. Advertisers are another matter. I have a Google Home Mini. It remains unplugged 100% of the time we're not using it. My phone can always go in another room. If we decide to turn it on to play with it (search something benign or play trivia) then we'll have it on. It's good for that kind of thing, and those are useful enough we don't mind letting Google have the little data that might produce.
chrischen 8y ago
Well, Apple doesn’t have the patent mentioned for a software control to spy on you.
- orbitingpluto 8y ago
  Apple did buy Shazam last December with its "always-on technology".
  - chrischen 8y ago
    Actualky Siri is already always on, and it’s opt-in. Shazam is only always on when the app is open. In any case having the capability to spy on you is fundamentally different from having the actual on-paper intent of doing so, such as a patent.
- menacingly 8y ago
  I see your point, but it's conflating a couple of issues. The comment I'm responding to was saying that a smartphone isn't a spy device because it doesn't listen to you, and the Echo does. My point is they're both spy devices, just opaquely crippled to varying degrees. To the other point, I'm not entirely sure I'm comfortable with the patents, but it's at least an Echo simply gleaning more from a voluntary engagement with it. I think the general feeling is an Echo is "spying" when it engages itself to listen to your home. Of course, I have no way to know if my echo is doing that, or if it were to start, and it's out of my hands to prevent it. The same with a smartphone. If Amazon wanted to spy, probably a smart tactic would be to deliberately "flaw" the wake word recognition such that it could plausibly be a bug, but the result is activating very often randomly. They already kind of do that, it would just need to be more frequent.
ryandrake 8y ago
There is a huge difference between “my phone is probably not listening to me” and “my home surveillance device, built for the sole purpose of recording as much as possible about me, in order to manipulate my buying habits—is definitely listening to me.”
- jhall1468 8y ago
  Your phone is always listening for the command "ok, Google" or "Siri" in the exact same way the echo is listening for "Alexa". So there's not a huge difference. They are the same thing.
  - ryandrake 8y ago
    You know how my phone is configured? Mine doesn’t respond to any of those phrases, so I conclude there is a small, but nonzero chance that it is listening to everything I do. Conversely, the sole purpose of an Echo is to listen to its environment and take actions based on what it hears. There is a high probability that it is listening at all times. You don’t see the difference?
    - jhall1468 8y ago
      No, because much like your phone the Alexa can be muted and entirely controlled from the very device you have deemed "safe". You don't need voice commands, they just happen to make it better than a typical bluetooth powered speaker. Still doesn't change the fact that placement of the Alexa is stationary, so your phone (regardless of how low the probabilities are) is still a considerably more "high risk" device in terms of what it could pick up than an Alexa. To me this just sounds like another person in a whole line of them that screams about privacy unless they have to give up something that's extremely valuable to them. The Alexa isn't valuable to you. Your phone sure is.
- menacingly 8y ago
  You've just persuasively reworded "both my phone and my echo are currently configured to not spy on me and the vendor assures me of this" At the base of the claim that Amazon is listening to what your echo can hear is the assumption that they're lying, and at the base of the claim that your iphone isn't listening to you is the assumption that Apple is not.
illvm 8y ago
I'm fairly certain it's baked into the chip or part of the firmware which is not easily writeable, which is why the wake words are static. One can check if a device is compromised or not behaving as expected fairly easily by monitoring network traffic.
- felix_nagaand 8y ago
  Obfuscating the data would be no more difficult than obfuscating tor traffic by delaying, stuffing, and merging packets. Has anyone done a long term packet analysis of an echo, home, iPhone, pixel, or galaxy?
- comex 8y ago
  As far as I know, the hotwords are static because the detector uses a neural network pretrained specifically for those words. Training a new network for a custom hotword would require way too much training data. While it would be possible to instead use a generic network that decodes any type of speech – like it does when in full speech-to-text mode – and then use the outputted representation to compare to a custom hotword, that would be far more compute and power intensive, whereas the specialized network is relatively small. (The specialized network is also probably more accurate.) At least on battery-powered devices, more power usage for something that‘s always running is a dealbreaker. The same pressure doesn’t necessarily apply to home devices that are always plugged in, but all of the major in-home voice assistants also run on phones, and the providers probably prioritize a uniform user experience across devices. Disclaimer: I could be completely wrong, and/or the situation could change over time due to improved hardware and software. There was a rumor earlier this year that the Google Assistant might add custom hotword support in the future: https://9to5google.com/2018/01/29/google-app-7-20-assistant-apk-insight-teardown/ https://9to5google.com/2018/01/29/google-app-7-20-assistant-... Main source: Apple’s article about “Hey Siri”: https://machinelearning.apple.com/2017/10/01/hey-siri.html https://machinelearning.apple.com/2017/10/01/hey-siri.html

_jal 8y ago

You also don't know that you're not being tailed by a PI, that your neighbors are not focusing a mic on your home, or that your dentist isn't on the lizard-peoples' payroll. However, we all make certain risk tradeoffs based on numerous things, including proxy signals. For instance, I would be intensely surprised to find out my downstairs neighbor was recoding me, based on multiple years of interactions. Likewise, I find the signal Apple sends ("We need a US-based GDPR") about privacy rather more compelling than the signal Amazon sends ("scour the mic for hints about product we can shift"). I mean, sure, while both $huge_bank and your hypothetical deadbeat relative might both steal your money, I bet I know which one you choose to to leave your money with.

menacingly 8y ago
I'm just saying it boils down to trust either way, not weighing in on which way that trust should lean.
52-6F-62 8y ago
> You also don't know that you're not being tailed by a PI, that your neighbours are not focusing a mic on your home, or that your dentist isn't on the lizard-peoples' payroll. Totally tangential, so feel free to disregard: This statement reminded me of my experience during my brother's TS clearance (Canada). CSIS readily had accurate data (in surprising detail) about my time spent during a short period they must have observed his family members to use during an interview with him. It was pretty fascinating, mildly disconcerting— but in this case didn't bother me (in the context, I understood why. I certainly wasn't interesting at the time). They only could have captured the data they had by a combination of tailing (however brief or otherwise more expensive much more technical means) and bank records. I was of course none the wiser. Advertisers are another matter. I have a Google Home Mini. It remains unplugged 100% of the time we're not using it. My phone can always go in another room. If we decide to turn it on to play with it (search something benign or play trivia) then we'll have it on. It's good for that kind of thing, and those are useful enough we don't mind letting Google have the little data that might produce.

chrischen 8y ago

Well, Apple doesn’t have the patent mentioned for a software control to spy on you.

orbitingpluto 8y ago
Apple did buy Shazam last December with its "always-on technology".
- chrischen 8y ago
  Actualky Siri is already always on, and it’s opt-in. Shazam is only always on when the app is open. In any case having the capability to spy on you is fundamentally different from having the actual on-paper intent of doing so, such as a patent.
menacingly 8y ago
I see your point, but it's conflating a couple of issues. The comment I'm responding to was saying that a smartphone isn't a spy device because it doesn't listen to you, and the Echo does. My point is they're both spy devices, just opaquely crippled to varying degrees. To the other point, I'm not entirely sure I'm comfortable with the patents, but it's at least an Echo simply gleaning more from a voluntary engagement with it. I think the general feeling is an Echo is "spying" when it engages itself to listen to your home. Of course, I have no way to know if my echo is doing that, or if it were to start, and it's out of my hands to prevent it. The same with a smartphone. If Amazon wanted to spy, probably a smart tactic would be to deliberately "flaw" the wake word recognition such that it could plausibly be a bug, but the result is activating very often randomly. They already kind of do that, it would just need to be more frequent.

ryandrake 8y ago

There is a huge difference between “my phone is probably not listening to me” and “my home surveillance device, built for the sole purpose of recording as much as possible about me, in order to manipulate my buying habits—is definitely listening to me.”

jhall1468 8y ago
Your phone is always listening for the command "ok, Google" or "Siri" in the exact same way the echo is listening for "Alexa". So there's not a huge difference. They are the same thing.
- ryandrake 8y ago
  You know how my phone is configured? Mine doesn’t respond to any of those phrases, so I conclude there is a small, but nonzero chance that it is listening to everything I do. Conversely, the sole purpose of an Echo is to listen to its environment and take actions based on what it hears. There is a high probability that it is listening at all times. You don’t see the difference?
  - jhall1468 8y ago
    No, because much like your phone the Alexa can be muted and entirely controlled from the very device you have deemed "safe". You don't need voice commands, they just happen to make it better than a typical bluetooth powered speaker. Still doesn't change the fact that placement of the Alexa is stationary, so your phone (regardless of how low the probabilities are) is still a considerably more "high risk" device in terms of what it could pick up than an Alexa. To me this just sounds like another person in a whole line of them that screams about privacy unless they have to give up something that's extremely valuable to them. The Alexa isn't valuable to you. Your phone sure is.
menacingly 8y ago
You've just persuasively reworded "both my phone and my echo are currently configured to not spy on me and the vendor assures me of this" At the base of the claim that Amazon is listening to what your echo can hear is the assumption that they're lying, and at the base of the claim that your iphone isn't listening to you is the assumption that Apple is not.

illvm 8y ago

I'm fairly certain it's baked into the chip or part of the firmware which is not easily writeable, which is why the wake words are static. One can check if a device is compromised or not behaving as expected fairly easily by monitoring network traffic.

felix_nagaand 8y ago
Obfuscating the data would be no more difficult than obfuscating tor traffic by delaying, stuffing, and merging packets. Has anyone done a long term packet analysis of an echo, home, iPhone, pixel, or galaxy?
comex 8y ago
As far as I know, the hotwords are static because the detector uses a neural network pretrained specifically for those words. Training a new network for a custom hotword would require way too much training data. While it would be possible to instead use a generic network that decodes any type of speech – like it does when in full speech-to-text mode – and then use the outputted representation to compare to a custom hotword, that would be far more compute and power intensive, whereas the specialized network is relatively small. (The specialized network is also probably more accurate.) At least on battery-powered devices, more power usage for something that‘s always running is a dealbreaker. The same pressure doesn’t necessarily apply to home devices that are always plugged in, but all of the major in-home voice assistants also run on phones, and the providers probably prioritize a uniform user experience across devices. Disclaimer: I could be completely wrong, and/or the situation could change over time due to improved hardware and software. There was a rumor earlier this year that the Google Assistant might add custom hotword support in the future: https://9to5google.com/2018/01/29/google-app-7-20-assistant-apk-insight-teardown/ https://9to5google.com/2018/01/29/google-app-7-20-assistant-... Main source: Apple’s article about “Hey Siri”: https://machinelearning.apple.com/2017/10/01/hey-siri.html https://machinelearning.apple.com/2017/10/01/hey-siri.html