I'm curious if the police will push this, as it could reveal how much voice information is actually stored. Amazon has declined to answer how much voice information is stored online[1] and how much if that is tied to your identity. We do know that Alexa stores a list of queries you've spoken, and you can "delete" these records, but it is unknown if this impacts the actual voice data behind them.
Wait a minute, we'd already know if Alexa was recording everything by the network traffic that obviously has already been sniffed. We can assume that Amazon is storing every "Alex,..." query that you've ever done. So what is there to know?
We know they are streaming it to do all the NLP in the cloud. This is obvious from the computational costs alone of doing NLP.
The important question here is "What happens to the audio data streamed to the cloud, a few seconds AFTER the query has been identified (or not identified) and the results sent back to the Alexa device?"
The nieve assumption is "all audio data is deleted after processing". The reality is that data is still valuable to Amazon for a variety of uses such as
(1) further training their voice recognition software
(2) advertising data mining [how many people are in the room, things they are talking about --- note, Facebook's mobile app infamously does this]
If they just store the query text, that is 'best case' from a privacy perspective. If they just store query text and query audio, that is less than ideal, but not too bad. If they store all audio, ever recorded, for an indefinite period of time... that is what this police request could reveal: Audio data stored for a non-keyword-trigger, and for days/weeks after the fact.
No need for speculation: you can open the Alexa app and play back your own queries, audio and all. There's even a help section titled, "How do I delete Alexa voice recordings", which reads:
"You can delete specific voice recordings associated with your account by going to History in Settings in the Alexa App, drilling down for a specific entry, and then tapping the delete button. Or, you can delete all voice recordings associated with your account for each of your Alexa-enabled products, by selecting the applicable product at the Manage Your Content and Devices page at www.amazon.com/mycd or contacting customer service."
Once more though: that's just the audio flagged as a "query" by Amazon and stored as such. What about other stuff that gets uploaded prior to NLP & validation?
It's a serious question, and one that I think Amazon (and anyone else streaming audio to the cloud, Siri and Google Voice Search are only slightly less constrained) needs to have an answer for.
It's a great question. Reminds me of Uber and its new policy of monitoring location of customers few minutes after drop off "to improve service".
It is also true that if Alexa was listening in on conversations so as to have "social conversation context" for "improved processing of commands" ... well.
> What about other stuff that gets uploaded prior to NLP & validation?
I would be very surprised if Amazon sends everything up to the cloud; it would be very expensive to do so. Looking for "Alexa" in the audio is done locally, and probably triggers recording then.
I don't think the expensive part is accurate. Voice sounds okay using the G.729 VOIP codec at 8kbps. Capturing and storing that amount of data, for every Echo user, 24 hrs a day would be trivial for company the size of Amazon.
As others have pointed out, network monitoring has shown the Echo only transmits after it's heard the wake-word, so it would appear Amazon doesn't capture everything.
Hypothetically, if I was tasked with recording everything, I'd just add an internal buffer and then ship chunks of that data along with the regular queries.
Indeed, yes, that'd be the way to do it. You could claim the post-wake-word data is recorded at a higher bit rate than it actually is, thereby accounting for the larger than necessary data-transfer.
I won't be surprised if we find out this is happening. They'll probably call it a "bug", fix it in an OTA, but then accidentally regress 6 months later.
Something you should probably understand about Amazon is that their Leadership Principles are real, and customer obsession is their main focus.
A big part of customer obsession is not eroding the trust of customers by treating their privacy as extremely important. For example, I've never heard of an incident in the more than two decades that Amazon.com has been in business of them selling customer data to a 3rd party for marketing purposes or otherwise. They wouldn't do something that could erode customer trust, or would be anti-customer, because they are in business for the long haul. Customer trust can only be earned slowly, over time, but can evaporate instantly with one mistake.
I trust Amazon more than Google and others to protect my customer data, because they've frankly earned this trust over 20+ years.
I won't be surprised if we find out this is happening. They'll probably call it a "bug", fix it in an OTA, but then accidentally regress 6 months later.
I'd be very surprised if this is happening -- Amazon knows people are watching, and it would erode trust with customers if it's found out that they are spying.
I'd be much more concerned about whether or not my phone is spying on me, even if it says it's only listening if I say "Ok Google" or "Siri". It's much harder for me to see what traffic my phone is sending across the cellular network than to snoop Amazon Echo's wifi traffic. And phone malware (sometimes baked in by the manufacturer) is increasingly common. Since the Echo does not allow user-installed Apps, at least it's less vulnerable to malware.
But you can't prove it's not happening. I'm personally skeptical that it is happening, but you saying "No for real guys, it isn't happening! I double-pinky swear" isn't actually proof of any kind either way.
I think you may be making a distinction where none exists. Can you give me an example of what "other stuff" you're referring to? (Disclosure: I worked at Amazon on a related product.)
As others are posting here, the Echo is obvious about when it's recording: when the onboard hardware detects the wake-word, the LED rings light up blue and it begins streaming audio to Amazon until the Alexa service determines that the query has been completed or that there's nothing more to hear. People have monitored the network activity of their Echoes to verify this (and if you don't trust that, well, nothing I say will help you.)
Sometimes the wake-word detection is mistakenly triggered and some unrelated audio gets streamed to Amazon, but the LEDs will still light up, and at times it'll even reply to a query that was never intended (I've been watching TV when my Echo will hear something and reply that I don't have any timers set, or whatever) but these recordings would still be covered by the link I pasted in the help text above.
I think a lot of the HN paranoia about these devices is overblown. All of this functionality is provided by devices that you already own, including hardware wake-word detection on all modern smartphones, which I assume everyone here carries around everywhere and recharges daily without much concern.
I think there's a stronger mental disconnect here in a lot of consumers. Detecting the wake word is easy to do locally; doing natural language processing on the resulting audio recording is what must be offloaded to Amazon, as it's too computationally heavy to perform that process locally.
Because the wake word is able to be processed offline, there is no need for the Echo device to record and transmit audio all the time. While the hardware is physically capable of doing this (and therefore, it will never be possible to fully eliminate doubt in the device's security) it would not be in Amazon's or the customer's best interest. Talk about a waste of resources, and a privacy nightmare!
But it could still record background audio and transmit it later to Amazon, when an Alexa query is triggered. How do we know that it doesn't do that? Has anyone opened up the data stream to check what audio it contains?
You can measure the amount of data it's sending so you can at least verify it's not uploading vastly more audio than expected.
They could do sneaky stuff like upload the audio after the wake word + a few seconds BEFORE the wake word. The cost benefit of this doesn't seem high especially if you consider that Amazon says all the time that they want to 'earn customer trust'.
Maybe they have hundreds of secret wake words triggering recording and covert upload of audio. It's not hard to implement - they already have 3 wake words, and ability to record audio locally. A list of national security words and marketing words would be the first choices that pop into my mind.
Do you trust them not to waste this opportunity to get more user info? What about future software updates and especially user targeted software updates? They could avoid detection by only doing surveillance on selected targets. When they are caught, they can justify it with some "improving services" or "software bug" bullshit excuse.
> doing natural language processing on the resulting audio recording is what must be offloaded to Amazon, as it's too computationally heavy to perform that process locally.
This is an unquestioned assumption actually.
I think it's way more to do with the NLP system being very very proprietary.
I wouldn't be surprised if it could be done locally, computing is pretty powerful, but they never bothered to even try and make it work locally. Because "cloud computing" is a dream come true for businesses, hiding their special sauce, not having to explain what they're doing (whether it's hard or not) and always maintain control.
Typical audio compression rates can be 10:1, especially for voice. So if the LED rings are lit for 30 seconds, is there 30 seconds of network activity, or 3 seconds? If it's 30 seconds, either Amazon is grossly wasting internet bandwidth by streaming uncompressed audio, or it is sending 10x as much data as the query it is supposed to be sending.
You do understand that it is trivial to operate the LED lights and microphone independently, right? The hardware /could/ be designed so that the microphone can't be activated without also turning on the LEDs, but I'm guessing they're independently controlled by a microcontroller, and that it's the hidden software which is responsible for turning the LEDs on when the microphone is in use.
Yes, this is a concern for smartphones as well. The capability, even if it is not currently in use, is a concern.
I'm completely with you. Recording video from webcams without turning on the light indicator has been exploited for years. [1] The FBI director admitted to putting tape over his webcam. Why would anyone think audio is any different?
And I'd like to add that even if Amazon doesn't do this, malware could, or certain state actors may compel Amazon to ship this on certain customer's devices.
Or if you have a conversation about coffee[1] including the word Alexa[2], but that cannot be parsed or validated as a "query" and produces no response from the device, does that audio get stored somewhere? Could it be captured in transit by other actors? Seems like an important distinction to me.
[1] Or, y'know, about some subject more subpoenable or privacy-sensitive.
[2] Or, y'know, not including the word Alexa but triggering the on-device recognition anyway. "Alex, uh, here's your payment for murder."
Sentences after the 'Alexa' key word are still recorded as un-identified questions that you can view in the Alexa app. Source: bought mom one, have used it.
According to others in the thread (i know nothing of this), it sounds like that is easily verifiable though, and not done. Ie, it sounds like they've monitored the outbound traffic of the device and it only sends Alexa... requests.
It's possible that they keep a rolling buffer of all audio in the device though. There could be legitimate reason to hold onto such a window of audio. Would make for an interesting discovery.
What would be the legitimate reason to do so? Queries are always preceded by the keyword, so what legitimate reason would there be to record and store audio before the keyword?
I could imagine trying to capture and analyze at least a few words leading up to the trigger word of 'Alexa' to help make it more accurate.
Even if the trigger word is Alexa, different languages, dialects, and slang might vary quite a bit. Some significant portion of people might use "Yo, Alexa...", vs "Hey Alexa...", vs. <a second or two pause> "Alexa...". I'm sure someone with a spouse or child named "Alex", "Alexander", "Alexandra" and so on might also trigger it by accident a bunch, e.g. "Alex, uh, who called?".
Matching all of those trigger patterns would help filter out false positive matches, or take false negatives and refine them so more people can use the product with their own language style and be less frustrated.
Allowing that sort of sentence construct violates the privacy afforded by starting with the key phrase. If the Echo did allow this, I would not own one since it would send a lot more unintended sentence fragments to Amazon.
By saying "Alexa" first, it gives me a chance to hear the wake-up tone before I continue my sentence.
So if I'm saying to my friend Alex "Alex uuh, tomorrow we will rob the bank", I have a chance to stop before incriminating myself. But if I say "Tomorrow we'll rob the bank Alex uh... are you in?" and the Echo interprets that as a command, then it's too late, it's already been sent and analyzed.
It's strange that you'd have concerns about privacy when it comes to sentence structure and not that you have a literal black box with a powerful computer and an always on internet connected microphone in it.
I trust Amazon more than Samsung, Chevy or, well, Samsung. But at least I can verify that my Echo is not sending home unusual amounts of data when it's not in use.
I think they're talking about the distinction between the actual audio, which can be listened to and deleted from the Alexa app, and the processed data breaking it down into a (potential) query, possibly sitting nowhere but some query logs for underlying services.
The 'Alexa' voice recognition is done on the client, not on the server. Once the trigger word has been spoken, the audio data begins streaming to the server.
They carefully worded it so that while it is true it only starts transmitting when there is a query, it doesn't stop transmitting afterwards. It could continue recording and sending to the server for the rest of the day, not just the query (just kidding.. I hope).
Ummm source? I know there has been a bunch of paranoid conspiracy theories about this that are mostly backed by confirmation bias, but I don't believe anyone has actually came out with any proof.
The speech recognition part could run on a Pentium 133 in the nineties (Nuance software, Dragon naturally speaking), the Echo device has probably a more powerful CPU than that.
The NLP part could run on the device as well, it's mainly a storage problem, you need dozens of GB, at least 32GB sdcard.
Ideally only the search phrases would be sent to Google/Amazon/etc. That's also how Bill Gates vision was described in his 1995 book The Road Ahead. A personal agent in a wallet PC would surf the information super highway for you. ...well we are "almost" there but with a different solution.
Having the data locally necessary to process the wake word is slightly more practical to havig the language corpus necessary to process all English language in hundreds of accents.
Speech recognition worked fine in the late 90s. You could optionally train it for 15min with your own voive to improve the results. The Nuance software was better back then than what shipped with Win Vista inbuilt by default in 2006 and later. The same Nuance software runs nowadays on servers to power Siri, Google Now, Cortana. And who in the right mind would think every user gets its on dedicated super-computer. In reality, cloud based speech recognition and NLP has the advantage of a central database to collect different voice samples (to train) and a multi Gigabyte central NLP database, etc. But you won't get more CPU cycles than what would be available on a modern smartphone or a Smart TV or (what would fit in) an Echo like device. An offline-available software assistant as kind of premium feature for people who care.
1) Put the speech to text.
2) Put the text to speech.
3) Send the data over Tor.
Although #3 would use a Tor exit node and would allow packet sniffing/logging/mangling etc, it'd at least be more anonymous.
Thing is, this is in nobody's interest. Not in the police or Amazon's interest. For the owner, it'd depend on if they are the defendant/suspect or the victim. The victim is not always the person who owns the Amazon Echo. But an Amazon Echo is a nice point of defense because the proof cannot be deleted anymore. All one has to say during a robbery is "Alexa, call the cops". That's quicker than having to dial the number. And at any point, if you do say Alexa, you can also trigger a warning or proof. Ultimately, if you could change the keyword to something more generic you could trigger it without a robber noticing, or by tricking them. A keyword like "put your hands up"? I don't know.
So we know that they have a record of every query that Alexa has detected. I'm guessing what you're questioning is whether they're storing the actual voices so that they can differentiate between different people that have queried Alexa.
I'm quite confident that they store the voice data. It's very valuable to them as a source of training data for future versions of 'Alexa'-like products.
"Hey we allow you to delete past recordings associated with your account and as far as you are concerned they will be, but we're gonna keep a copy for ourselves right over here, that you'll never know about."
Yes, they are storing it. Source: Amazon web site and Alexa app (which let you review / listen do / delete said audio). They are very up-front about this.
The device doesn't start listening when you say the wake word, the device is always listening. It's only the UX that responds when the presence of the wake word is detected. This means that the device is continuously monitoring the environment for the presence of the wake word we know about ("Alexa").
Can the device monitor for other words that trigger other UX sequences? Of course it can!
Can Amazon deliver an OTA update to this network connected device to change behavior after it leaves the factory? There is no reason to believe that they cannot.
So, no, I don't believe the police are wrong here. In fact, I would be in favor of law enforcement having the ability to "tap" these network-attached microphone appliances with a warrant. Then, perhaps, people will begin to understand what they've really bought.
EDIT: Passive monitoring ("listening") is not the same thing as storage ("recording"). In some cases below, these are being conflated.
1) The device is storing any amount of locally recorded audio. So far evidence says no.
2) Amazon would OTA all echos to upload everything it hears. So far evidence says this hasn't happened. It also likely won't because IMO security people monitor these devices more than anything else on the planet. Some I believe for good reasons, the others just want to play gotcha.
3) If 1 and 2 are true, Amazon keeps the data for an extended period of time.
Can we please not tinfoil hat the hell out of these devices like we do the rest of IoT sometimes?
(I think) the argument your parent is making is that Amazon could easily send a software update to a specific Echo to make it record everything it hears.
If at the request of the government, yeah that's a problem unless there is some kind of warrant, right?
I'm all for people being suspicious, but the intent of this was to say "the cops are hoping it might catch something" as in it was accidentally triggered.
Yes and that's a problem with law not catching up with technology. Unfortunately for many people in the country, other social/financial issues are more important to them.
> In fact, I would be in favor of law enforcement having the ability to "tap" these network-attached microphone appliances with a warrant. Then, perhaps, people will begin to understand what they've really bought.
All your arguments apply equally to any smart phone. It's another network attached microphone. Are you in favor of it there too?
Why not? Seems like a subpoenad record would serve as an effective canary for determining what end user data is actually stored. Taken to the next step, LEA should be able to ask google alone for your browsing history for the last 10 years based on Google Analytics. Storing customer data should be seen as a corporate liability if you're not upfront about what is stored, used and how.
But when you bust the door the owner knows you did that. It's more insidious when they get in covertly to install microphones, search the house and plant substances.
It's certainly not equivalent. "Third-party doctrine" is extremely bad law that ignores the notion of private relationship. The law even recognizes some notions of private conversations (with your lawyer, and with your doctor), but ignores others (with your cloud provider) because the law is older than the cloud and the government loves making power grabs.
Justice Sonia Sotomayor is right to challenge it. I fear that she is one of a narrow generation that is young enough to understand digital technology, and old enough to remember the right to privacy in personal effects.
Third-party doctrine is not actual law, and challenges only exist as 4th Amendment cases on specific uses (i.e. not the doctrine as a whole). While Sotomayor is right to raise the issue, sadly her words don't constitute action on the topic.
>Wouldn't you want the police be able to intervene when they hear or observe shouting or fighting through an open door or window?
Different things. In a closed room, there is expectation of privacy. It is not the same for what can be seen from an open window.
Also, there is no assurance that LE won't misuse this to crush any dissent or silence somebody they don't like. It is easy to take something said at home out of context or doctor it to sound like something else.
I think the media have made it clear to the public that there's no expectation of privacy around these devices, and that there would be a hard case to make in court that someone was so insulated from the media that they wouldn't have any knowledge of what those products could be enabled to do with full, legal collaboration between law enforcement and the manufacturer.
This argument will not stand in the courts. Data collection without consent is illegal, which is why there is so much controversy around the stingray things. Getting a warrant to look at the data may be legal, but collecting the said data without consent is illegal in the first place.
Alexa listens only to the words after the wake word. What's more it doesn't capture more than 10-12s past the wake word. If you have a device you can check this yourself (Alexa, One Mississippi, Two Mississippi…).
It's actually hard to get it to capture this much. Take a breath or pause to long and capture stops sooner.
The request of the police is nutty and indicative of not understanding the technology. Unless the murdered woman yelled out "Alexa {pause a little bit for wakeword recognition} Heeelp, my husband is killing me, call the police" or the husband asked "Alexa, how do I clean a bloody hottub", nothing material will be discovered.
I have 4 of them. After the first one I checked what was captured, what was stored, what was streamed. Nothing that mattered.
Checking this by saying "Alexa, <long string of words>" and seeing at what point it starts doing something does not prove that it's not listening all the time, doesn't prove that it doesn't keep recording, and in general only tells you about Alexa's UX, not the underlying technology.
It stops listening. No action required. You can check in the Alexa app what is recorded (I just did). You can also check what is sent out to Amazon (which is what I did when I got the first one, using wireshark). The hardware is listening all the time, for the wake word. Nothing is sent until a wakeword is received.
True, it does have an 8GB flash drive, according to iFixit (4GB in the Echo Dot). For now, I'm giving it the benefit of the doubt. My cell phone is a much more useful target for the government (location, video/audio, browser history).
That's exactly what Amazon would like there users to think. How do you think he wake work is detected? It MUST be listening to some sort of ambient audio, wether that is recorded or kept in anyway, only Amazon would know.
But that won't give you the whole story. They could batch up the network traffic to send later, or push an update that changes how much they listen after your traffic monitoring is done.
>They could batch up the network traffic to send later,
I was thinking that too, but realistically, someone would eventually discover this (consider how much outbound traffic would have to go at once in order to store that much audio).
>push an update that changes how much they listen after your traffic monitoring is done.
Again, someone would eventually discover this. It wouldn't be hard to constantly monitor network traffic for noticeable discrepancies in outbound activity.
Furthermore, a discovery like this would not go overlooked by the general public. Amazon would be doing a serious disservice to their reputation by going about extracting the audio in such a mischievous manner.
I can't say I agree. How long did it take for us to figure out that certain camera's were sending photos back to china? or that other apps do things that seem malicious. Many people watch those kinds of stats, but the vast majority of them don't.
Additionally, just because they have it always recorind for one person, doesn't mean they have the same software running for everyone else.
TheInformation's article includes a bit more info and this tidbit:
> As most owners probably have experienced, the microphones can often be triggered inadvertently. And those errant recordings, like ambient sound or partial conversations, are sent to Amazon’s servers just like any other.
When I use my Echo in my apartment alone I don't have a lot of false positives on the wake-word detector. This past week I've been in my family's home and very frequently when more than one person is talking Alexa has butted in (and tried its hardest to understand what we're saying, listening for the maximum time it can).
I think the hope is that if some ruckus was happening in the home, the Echo might've accidentally woken up at some point and recorded a small bit they can use to help piece together the crime.
Absolutely right. The people here gleefully, cynically dismissing this as police stupidity would be demanding the police investigate even the smallest lead to help get justice for the murder of a loved one. If there's even a small chance the always-activated audio recording device in someone's home was inadvertently activated in the commotion, why wouldn't the police investigate it as an option? Even if there's not a recording of "help amazon my husband is murdering me," what was captured may help verify other evidence or set a timeline. Hell, even being able to prove the crime had or had not taken place at a certain hour based on whether Alexa was being used for its usual purpose is potential evidence.
I'm not sure you and I have the same definition of "listening". It must listen to hear the wake word, and likewise must process incoming sounds to do so. If you know of a way to do that without storing the incoming sounds, even if just briefly in temporary memory, I'm all ears...
>"The request of the police is nutty and indicative of not understanding the technology. Unless the murdered woman yelled out "Alexa {pause a little bit for wakeword recognition} Heeelp, my husband is killing me, call the police" or the husband asked "Alexa, how do I clean a bloody hottub", nothing material will be discovered."
The police often check search histories in murder investigations. Do you also suppose that is only useful if the victim googled "Help, my husband is murdering me!" ?
Maybe you should leave the policing to the police.
Alexa is the service in the cloud - Alexa doesn't get the words that are spoken before the device (the Echo / Dot) detects the wake word.*
This sounds like a small distinction, but it is critical to a clear discussion of what's going on.
If you review the interaction history in the Alexa app or on Amazon's web site, you won't see those utterances that did not result in detection of the wake word, because that audio was not streamed to the cloud.
* Technically, there is a very short buffer of audio that is captured on a rolling basis in order to detect and stream the front of the wake word sound.
I think the thing they are hoping for is accidental invocations, I've had instances in the past where something on the TV has 'awoken' the echo, even without saying 'Alexa'
Sure, maybe they don't understand the technology, but why not just hand it over so they can determine for themselves that there's no data pertaining to the investigation. They're essentially obstructing justice. They're trying to act like the good guy, but they're really just being assholes.
Of course there's the possibility that they have something to hide and are actually collecting more data than we all think they are.
How are they obstructing justice by simply asking for a warrant? Should they simply divulge private recordings to any random cops that ask for it? Is that how you wish companies treat your data?
The request is not nutty, perhaps the victim was aware enough to ask "Alexa, Call the Police!". FWIW This could be a great replacement to those fallen and I can't get up buttons for the elderly.
Given Stuxnet, your position seems dangerously naive.
There is really only one question: can this device be remotely modified to passively monitor a specific device in a way that is relatively undetectable to the end user?
If the answer is no, then fine. If the answer is yes, then what guarantee do you have that the ratchet of tyranny will not eventually deploy those capabilities? "I'm not hiding anything" is a poor response.
Am I saying people shouldn't be skeptical? No. But I'm saying people are focusing on a subset of devices when they've had cordless phones, cell phones, and PCs in their homes for decades.
I just get tired of the constant fear-mongering, if we have evidence of a problem, call out Amazon and have them fix it.
The cops in this story are banking mostly on the fact that it was accidentally triggered and they have some kind of audio, that's legit. Yet this article was written as if Amazon might magically expose some recording capability.
PCs, cell phones, and cordless phones have all previously been tapped or hacked on behalf of law enforcement and other actors, so people generally know and understand what they're getting into there.
These microphone devices are new. People, even HNers, mistakenly think that these devices are not listening until a magic word is uttered (full stop). Many people will not realize that the connection to the internet is bidirectional for data AND code (updates). That's really bad - it means that people don't understand the capabilities of the device they've brought into their homes.
There is a problem here and the evidence is plain: (a) wide spread ignorance about how these things work, and (b) the hot mic combined with the network connection imply a capability, even if currently unexpressed. Device manufactures are not adequately informing the public of the possible side effects of these products.
Therefore, I believe that LE should gain access to these devices with a warrant under the Third Party Doctrine, and that these warrants should be published on sites like HN so that people may become aware just as they are aware of taps on phones.
I think the previous posters point is that the risk incurred by an Amazon Echo is the exact same as any modern smartphone. That iPhones and Android devices have supported Ok Google and Hey Siri for a while now, are also already always listening, and already get OTA updates.
Your claim that "these microphone devices are new" seems to imply you're excluding almost every smart phone we already have in our pocket, with the exact same network attached hot mic concerns.
I don't care if it is or isn't listening; it can, listen, if its owners want it to, because they control the software it runs. I can't do anything to control that save refuse to own one, so that's what I'll continue to do.
It just said to call the police, but given it can also turn the tree lights on and off it's not far fetched to imagine law enforcement being looped in via some nutty integration.
From the privacy perspective, IMO, once you put a device in your home that you know is always listening, and is forwarding an unknown amount of audio from your home to a third party, at that point, you have surrendered your right to privacy.
No, you haven't, any more than when you buy a cell phone with a camera and microphone built in. People still have rights regardless of the products they buy.
Actually, according to the third party doctrine, you have. The difference between this and a cellphone is that you voluntarily give up this information whereas a cellphone does not transmit such data voluntarily. There are surveillance laws and other problems if a phone is recording you unknowingly. With Alexa, the recordings are submitted voluntarily. I'd be shocked if the police don't get their hands on this data. It's not like Amazon can claim, like Apple, that they can't themselves access it because it's encrypted.
'The third-party doctrine is a United States legal theory that holds that people who voluntarily give information to third parties—such as banks, phone companies, internet service providers (ISPs), and e-mail servers—have "no reasonable expectation of privacy."'
Except that the device (service) is not streaming / transmitting your audio except when you indicate to it - in exactly the same way as you would with "Hey Siri" or "Ok Google" - that you intend it to do so. (Or if it makes a mistake and falsely detects the wake word; again, something that your mobile phone is also capable of doing.)
It's not. I think the poster may be confused about how the Echo works - it's not constantly streaming audio or transcripts back to Amazon, just when it's directed to do so via the "Alexa" trigger word.
You are absolutely not consenting to send full transcripts of your conversations to Amazon. If they are in possession of it, it's entirely illegal, and the third party doctrine doesn't apply.
Whether the device constantly streams data or sends it every once in awhile is irrelevant. In the agreement with Amazon you agree to give them them the data, including the voice data. Amazon even lets you manage/delete the data after the fact. They're not collecting it illegally. Whether it is full voice transcripts or not that you're agreeing to send to Amazon is not 100% clear, but that's also irrelevant because by using Alexa you've agreed to send that data. It really depends on Amazon's definition of "when you interact with Alexa." I'm pretty sure that's quite broad.
This isn't Apple fighting encryption backdoors. This is Amazon impeding justice.
"1.3 Voice Services. You control Alexa with your voice. Alexa streams audio to the cloud when you interact with Alexa. Alexa processes and retains your voice input and other information, such as your music playlists and your Alexa to-do and shopping lists, in the cloud to respond to your requests and improve our services."
I don't believe they are saying a EULA or ToS is "above the law" they're saying that there is a reasonable argument to say that the expectation of privacy was waived by the guilty party owning an Amazon Echo. If the ToS states these facts more clearly, then that argument would either have more or less weight. Without anything being said in the ToS or EULA, the argument could really go either way, depending on many things, like the judge's mood or interpretation.
But laptops, and smartphones have come with cameras and integrated microphones for a long time, the only difference to the Echo is that they don't claim to be always listening.
which means if a smartphone records you, you can play the "why was it listening?" card
with an echo, you can't claim "why was it listening!?" because you knew it was and that was the reason you bought it.
granted, I dont think Amazon is doing anything nefarious at this point (other than what they normally do, building an advertising repertoire about you). Also, most people seem to agree that it only transmits data to Amazon after being triggered. but thats not to say they couldnt start listening to everything in the future to "improve the product experience". And if you think they are just using the data to answer your immediate question and not to build a profile on you, youre in denial. A smartphone company has no reason to build a profile on you. An advertising/sales company does.
Depending on the brand of phone and the software you have installed on it. And for the phone to keep listening when you explicitly disable that feature would be an issue. Whereas pulling the plug on the echo is the only way to disable listening.
Siri isn't the main selling point of buying an iphone. cellular communications (data, sms, or voice) are. Whereas Alexa is 100% the reason to buy an echo. You can't really make the claim you didn't know the echo was listening (since it was the reason you got it), but a person could reasonably make the claim they bought a phone without realizing it was more than just a phone.
Mobile phones (well, at least iPhones and Android phones) are "always listening" for their wake words in exactly the same way that the Echo is "always listening." They are doing local signal processing to detect a wake word (phrase) and, upon detection, they stream audio to cloud-based service that performs ASR and NLP on it in order to perform some service on behalf of the user.
Yep, I'm amazed at myself on how much privacy I've given up for convenience from 10 years ago. As long as you know the trade-offs...
I've started smart-devicing my home and a little wary about that, not from a privacy perspective, but from a hacking perspective. I can almost imagine a sci-fi/horror movie there somewhere.
Alexa only waits for the wake word "Alexa." It would cost more money than God in hardware to store every thing Alexa ever heard. She saves you're requests. And shows you those request in the Alexa App. And you can also delete it if you wish.
There is no data for the police to have, because beyond requests, there is no data.
Unless someone knows more about this than Amazon is telling us?
It would cost about $30m a year if you tailor the system to flagging specific data for storage and don't naively store every moment (e.g. you scrap silent moments and use VBR encoding).
Storing a year's worth of 96kbps audio costs 380GB. If you don't record silence and you assume the people around an Alexa are only speaking for at most 4 hours a day on average, that goes down to 76GB a year.
So if you then assume 5m Alexa's are active at any given point in time that works out to 380k PB. Ok, that doesn't work yet.
However, if you then layer on a flagging system, where only certain users' full record is stored, or only "suspicious incidents" are stored, and you get this down to only flagging 0.1% of all data, you arrive at 380PB of storage.
Amazon Glacier costs about $88.000 a year per PB, but there's a profit margin included in that, so I'll assume it costs Amazon just $75k a year.
In conclusion, it would cost Amazon about $28.5m a year to run such a system. That's certainly within the realm of possibility and of what LE/SIGINT clients would pay; I assume the NSA would gladly pay that sum x100 for that capability. Sounds like it'd be booming business for Amazon.
I think you're orders of magnitude off in marginal cost estimates for glacier users. Datacenters are being built out for a small number of commercial users (e.g. Amazon's core business) and the size of modern HDDs would lead me to estimate that storage is about free in a modern datacenter, the scarce resource is disk-time for read/write operations. That is, projects like glacier let Amazon sell disk that would have otherwise been stranded.
It is also the case that a consumer level service like glacier presumably has more redundancy than what might be needed for best-effort storage of these recordings, where losing any fraction of them wouldn't really be a problem.
I'm not in the datacenter business, so I've been conservative for lack of experience with storage at PB scale.
I've chosen to err on the side of estimating it to be more expensive, because I think that makes the end result more convincing:
30m is chump change for parties like Amazon, and in reality it'll cost significantly less. 1m might well do. Maybe it's less still. You could combine flagging users with flagging low-certainty or keyword-containing transcriptions.
Either way, you don't need collusion with intelligence parties, just an unscrupulous or naive exac at Amazon that thinks the data might be worth a lot for training future learning models. Of course the more sinister but legal reselling to government agencies is a financially attractive option as well.
I really like the math here, but isn't this a bit pointless? The system wants to parse meaning from audio; storing just the text it parsed is a lot smaller. Store just the text and whatever machine learning score of how probable the text is correctly parsed and that sounds like something prosecutors would love to bring into court: "Please read this line and let's see what score you get . . . "
For improvements they'd store the raw input so that when a mistake happens they can manually try to figure out why the machine got it wrong (e.g. a hi-hat was hit while they were saying "deuce" so it sounded like "douche").
It could also store compressed voice waveforms in such a way that any reproduction from the compressed data would sound horrible but would be at least somewhat intelligible to human listeners.
1200 bits per second is almost enough for toll-quality speech -- and I'm referring to the state of the art a few years ago. Speech codecs are probably better now. But let's stick with 1200 bps. That's enough to store continuous speech in the vicinity of the device for a year, using only about 5 GB.
My guess is that if you cared only about intelligibility and not fidelity, you could do the job with 10%-20% of that space.
So yes: Alexa could easily be collecting and storing a vast amount of data that isn't immediately transmitted or used.
Based on the costs of disks alone, their cost per PB is actually around $35k. They likely get a volume discount, so we can lower that estimate even more and say $25-30k. Bandwidth is essentially free even though they charge ridiculous amounts of money for it on their services. You can get a 10Gbps link for as low as $2,000 if you buy it on-net and in bulk, Amazon probably gets it even cheaper. So ~3PB/month for $1,000/month.
I thought it was HE, but it must have been someone else that had a 10Gig deal for $2,000. Either way, that's for a single 10Gig link. If you're buying 100Gb - 1Tbps like Amazon is, you're probably getting an even better deal.
We just signed a contract with Level 3 for slightly more than the price you mentioned, but they had to build into us; which costs them ~$120k out of pocket, thus the higher price.
What about abduction cases, inside trading, tax fraud, drug and human smuggling? There it could help to have data from months ago, so any newly discovered targets instantly come with a bunch of evidence.
> It would cost more money than God in hardware to store every thing Alexa ever heard
Depends, first of all storing compressed audio isn't that space-expensive, especially in some long term data storage like s3. Additionally they could only be storing the transcriptions, but not the voice behind them, which would be a lot less data.
We don't know as Amazon hasn't been very forthcoming about the privacy aspects of Alexa. I personally suspect they are keeping some voice information so they can use it to improve their NLP. I hope they are doing so in a way that is detached from accounts / IDs, but you never know.
Additionally, you can indeed delete a record of the query from the app, but who knows if the voice data or even the query itself is still stored after deletion, just not visible to us end users.
If it stored everything (and not just requests after the watch word) then it would end up trying to store audio or transcriptions of so many hours of tv and random conversations that it would be ridiculous. And that's just my house. I imagine most people have one somewhere near a TV, and it would do the same.
I'll point out Facebook is using this always on recording for advertising purposes and one of those is to fuel a nielsen-like TV/movie/audio popularity business.
Basically, Facebook's always-on audio listening on their mobile app (Messenger I believe, but might be both these days) was giving this data. I can't remember the name of the company, but here is another tech company doing the same:
> Symphony uses just one: an app, downloaded to the cellphones of its more than 15,000 panelists. Audio recognition software then picks up whatever people are tuning into, wherever they’re tuning into it: their TV sets, their laptops, or their smartphones. “[It] measures everything you want to measure from one approach,” says Bill Harvey, a media research consultant who’s worked with Symphony
I would think it would be possible and even beneficial to dedupe the data (15m homes x NFL broadcast, for example). Link a list of each echo's text conversion given similar data but perhaps different background noise. Or maybe getting data from multiple echos in different homes at the same time allows for "noise" filtering (people asking different things while the same background noise is present).
In fact that's just 40gb per year, which is pretty doable even on a local SD card that Alexa could be fitted with. Or even if it stores one month and deletes the oldest one every day, that's still 3gb. Very doable.
It is also feasible the device could just wait and transmit only voice like data and drop other sound data... (Even my crummy baby monitor can detect the difference.)
Speaking of which I wonder what the net traffic usage of the Echo is?
Especially since Alexa already converts it to text! It wouldn't be outside the realms of possibility.
Would it be possible to test this? Check the battery life of the Dot in a completely silent room vs the battery life of a Dot listening to an audiobook played on repeat. If it is actually listening and transcribing it should have a higher power consumption and thus die faster - right?
For NLP research, you'd want something that preserved more information than text.
Questions for anyone in the field: how much is preserved? Is there a < audio but > text form that allows for iterative testing? Maybe the output of a first-pass pheneme decoder? If so, what kind of space requirements?
> Speech can be encoded with less than 10Kbps [1], which means a maximum of 108MB of data per day.
People only speak a few hours per day and "interesting" conversations could be sampled from time to time and some Alexa stations flagged for full upload, if they want to know.
Average man uses 15,669 words/day, woman uses 16,215. Let's say average household has two people and half those words are spoken at home, 31,884/2 = 15,942.
Let's say 8 bytes per word on average, just under
128k/day. That's a little under 47 MB/year. Not too expensive.
> There has never been any "study" showing that "women talk almost three times as much as men", although this non-existent "research" has been cited by dozens of science writers, relationship counselors, celebrity preachers, and other people in the habit of claiming non-existent authoritative support for their personal impressions;
Well shit my bad. That was the top result from Google with just those numbers quoted when I searched for average number of words spoken per day. I should've been more thorough. Let me see if I can find better numbers.
Unless the guy said "Alexa, how do I get rid of a body?" or "Alexa, play 'The Sound of Silence'" just after the time of death, I think you are right.
I could imagine some sort of log data being used to refute an alibi but what is implied by what is missing from the article, that it could be used as an after the fact witness, is not really feasible.
Even if Amazon doesn't store all audio by default now, I would if it could become a thing(if it's not already) for law enforcement to basically get a warrant for Amazon(and Google and everyone else who gets into this space) to store everything and send it to law enforcement for specific individuals.
A year of audio at 32kbps (more than enough for speech) is only 120GB. That's only a couple bucks of month on S3 -- at retail pricing. So it's technically and financially feasible to store everything if Amazon wanted to.
The part that listens for the "Alexa" word is analog not digital.
It cannot be updated remotely by firmware.
Crack open the device, test the analog component and confirm the signal only fires when you speak alexa and nothing else.
Having said all that, I wouldn't buy it. NSA is known to temper with devices behind the back of companies (no matter how trustworthy you think Amazon/Google/Apple are). Check what they did with Cisco equipment sold to foreigners. They intercepted the shipment, tempered with the device and no one was any wiser until the Snowden leaks. Cisco now ships via proxy to guarantee no tempering.
American businesses shouldn't fear globalization or chinese disregard for intellectual property; their own government is fucking them.
> The part that listens for the "Alexa" word is analog not digital.
> It cannot be updated remotely by firmware.
I used to think that about my phone, that it was hard-wired to "OK, Google". I had a Moto X (first generation), which is one of the first phones to have this always-listening feature. They claimed that it had special hardware that allowed it to work using very little power.
But after a software update, they added a feature that allowed me to retrain the activation phrase to whatever I wanted. They recommend using something with 5 syllables or more. It works. I haven't addressed my phone with "OK, Google" for years.
If Motorola (owned by Google at the time) can do it with Moto X, I'm sure Amazon could do it with Alexa too.
Every smart phone ever can do this. You have no way to verify the mic isn't in use, you have no way to verify cellular data transmission is not taking place, and you have no way to know the modem is not doing a remote execution injection to change whatever aftermarket free software OS you have running on the device is doing.
No, it means that you're sitting in your house without a roof, and saying "I would never forget to close my windows during a rainstorm." Never mind the rain is pouring through your roof.
You're seriously willing to carry smartphones, which have both microphones and cameras, and give no indication visible or otherwise, of when they are listening or recording or sending data, and run whatever dodgy code some asian device manufacturer wanted it to run, but you won't trust a device made by Amazon that people have run wireshark captures against and lights up with big blue LEDs whenever it is listening.
In defense of the downvotes (albeit its bad manners whoever is downvoting you), camera / mic LEDs are nonsense. They are all software controlled, and the control software is almost always proprietary, and thus you cannot trust those either.
Additionally, I'm pretty sure the Amazon Echo is just rebranded third party mic / soc boards put together by Amazon. It is still "asian device manufacturer" firmware.
> No, it means that you're sitting in your house without a roof, and saying "I would never forget to close my windows during a rainstorm." Never mind the rain is pouring through your roof.
I don't disagree. If there was a legitimate alternative, I would use that, but there isn't. I do everything I can to make sure it runs as much FOSS as it can.
> Reality called - it wants the tinfoil hat back.
How can you claim that attitude? You just admitted we have 2 giant vectors for attack/compromise, then dismiss them without any reasoning?
His point is that no one is concerned about the always on hot mic in their pockets, but it's the only thing people talk about with the Echo. Why aren't people claiming that everything you say is recorded on your iPhone or android phone? Or that it's all sent to apple/google? Yet they feel justified in claiming that of the Echo?
> Why aren't people claiming that everything you say is recorded on your iPhone or android phone? Or that it's all sent to apple/google?
uh... there's tons of people complaining about that all the time. I don't know about you, but that's a constant criticism of both companies that I see.
I've seen it, but not every time there's a discussion about Android or iOS. Maybe it's just because the always on mic is the sole feature of the Echo, while it's just a feature for phones. That said I feel like a lot of people do criticize the Echo while not realizing the phone in their pocket is capable of the same thing. Especially people saying they wouldn't buy an Echo or put one in their house because of privacy concerns, while they surely have an iOS or Android device on their person.
>Police say Bates had several other smart home devices, including a water meter. That piece of tech shows that 140 gallons of water were used between 1AM and 3AM the night Collins was found dead in Bates' hot tub. Investigators allege the water was used to wash away evidence of what happened off of the patio.
When every house is IoT connected, we will be very close to a 'post-crime' world.
This brings up an interesting question. If I say "Alexa, call the police!" or "Alexa, I need an ambulance!" will anything happen? It doesn't officially say it can be used to request emergency response and I'm scared to try it.
I wonder if they've received a lot of these requests and/or if they have an emergency team on standby.
I don't think anything will happen, although people can write their own "plugins". It's already a speaker/microphone, why not make it a speakerphone; if you slipped and broke something and can't even crawl to the phone, saying "Alexa, call an ambulance!" and being able to communicate with the rescuers would be helpful. Sadly it's not in their spec to be used as that, yay for closed hardware!
Alternatively, Amazon should have a log of music you asked it to play. The suspect is named Bates. If you were being chased by him, could you yell "Alexa, play music by The Police!" followed with "Alexa, play the soundtrack of Bates Motel!"? You'd be dead but at least there'll be a hint for the police somewhere in the cloud.
This would be of interest for the elderly, "help, I can't get up" with an omnidirectional microphone instead of a fragile, not waterproof, awkward beeper that you carry everywhere
Interesting that the police is going after this electronic speech data, reminds me of the instance where a parrot witnessed a murder that was posted on this site couple of months ago, in that case the parrot was not able to testify...
Also interesting that the lawyer is going for the reasonable expectation of privacy regarding that data, if that holds, then it could allow them to argue due protections from the 4th amendment.
But let's ask the question here; is data regarding usage of utilities private? It seems that something that you installed for water usage, should be. Could this data not have been acquired from the water department?
Late to the game here and apologize if this has already been mentioned but may shed some light on what is actually stored. A while back facebook officially announced what they "store" in regards to what they record when the app is open and has access to your microphone. The app itself turns your voice audio into a profile that is then sent to the server, it does not send your actual audio. This light(er) weight audio profile is then matched to words which are then used for marketing. I wouldn't be surprised if Echo, Google Home, etc do something similar since sending the full audio recordings would be both bandwidth and storage heavy.
"Audio profile" meaning MFCC coefficients? That's still enough to fully reconstruct a meaningful audio stream. Obviously, since it's enough to do speech recognition.
>"A ring on the top of the device turns blue to give a visual indication that audio is being recorded. Those clips, or "utterances" as the company calls them, are stored in the cloud until a customer deletes them either individually or all at once."
Is there a technical reason for storing this after NLP processing has been completed on the "utterance."? Wouldn't an ephemeral cache be sufficient?
I see that a lot of people say that police did that because it is incompetent. But what if this is actually a way for police and other low tier agencies to check how much Amazon collects? I bet NSA would never work with police and most likely not much with FBI etc. So this might be just a check? This is just a speculation, but we should not assume stupidity of some of the agencies just because its an easy answer.
[1] https://www.amazon.com/gp/help/customer/display.html?nodeId=... and https://www.washingtonpost.com/news/the-switch/wp/2014/11/11...