Activate this ‘bracelet of silence,’ and Alexa can’t eavesdrop
nytimes.com
nytimes.com
I can't bring myself to blame my fellow peasants for trying to take back some control.
The air we breathe in and communicate with over radio waves and audio waves is a public commons. Don't piss in the well.
If the only way to drop out is by using these, they are gonna get used. Whose fault is that? I blame the people who made it the only way to drop out.
That’s not how responsibility works. If you use this device, you’re responsible for the collateral damage it produces. The Trackers are only responsible for making you think you need it
It's not cool to make the most vulnerable people pay the cost of your protest.
The fact that you quoted the guy in the position of power to make your point is pretty ironic.
You’re not impacting the companies they perform this surveillance.
You’re not impacting any individuals other than the minorities that need these devices to function in a society that is already inherently biased against them.
So basically you’re protesting the existence of deaf people. Good job.
Don’t go jamming all the radios because you personally don’t like the uses that some radios are put to.
It would also be incredibly illegal
Wouldn't surprise me either way.
When the whole point of the critical system is to connect things wirelessly... your point makes little sense. There are backups, of course, but why would you go out of your way to make things worse when there's no fix and people already know it's an issue? What about that is tempting? It's a fundamental technological limitation - there is no jam proof wireless communication system.
Really? Causing temporary, localized connectivity outages, that's an inconvenience at best.
If the goal is to come up with something to defeat privacy invasion—and i think it should be–this isn’t the answer.
When did we as a society give up our privacy for such a small amount of convenience? Or has it been gradually chipped away until we have reached this point?
I feel sorry for my children and the future in which they are going to live in. I also feel a sense of shame for participating in a society that has normalised these things.
Sorry kids. We got greedy. We boiled the planet and normalised blanket spying for profits.
Read it on a Mozilla blog a few years ago and I think they quoted it too from somewhere else.
But I feel like it sums the whole thing up nicely.
For many, it does not feel like somebody is watching us. Just that data about us is dumped in a huge archive too vast for anyone to dig around in. This (Wrong) sentiment makes many assume that the currency is probably cheap.
But because Alexa isn't eavesdropping on us and believing this is the tech equivalent of antivaxxers: "If you don't believe it, you're part of the conspiracy"
Also, do you recall the fix? They remotely changed the activation phrase that triggered the laugh.
Do you recall the Google home devices with a minor physical flaw that caused the pucks to record and transmit everything?
Do you recall how these recordings are given to third parties to evaluate to improve the quality of the service?
I guess what I’m trying to say is that we don’t know when or if these devices are recording us. Yes, the same is true with our phones as well (though this bracelet from the article would work on them too). It might not be out of malice, but its safe to say that we as the customers have no real idea of how much they hear.
I don't see that as a big deal as the data is likely always sent to a remote server for parsing and processing.
Also, the firmware on these devices is remotely updatable. Even when you can't today, you may be able to tomorrow.
I'd be far more worried about the invasion of privacy caused by it listening to everyone's shopping lists.
https://www.theverge.com/2013/12/29/5253226/nsa-cia-fbi-lapt...
They could also compel Amazon to install an update that makes your Alexa record everything.
That doesn’t make vaccines bad though.
That's what they want you to think.
/s
This is surely not actually effective against a wide range of well designed, real world microphones, and definitely not against something designed to surreptitiously record you. Strikes me as more of a gimmick than anything else.
Full disclosure: I don't have any always-on microphones in my house, and probably never will, so I probably qualify as some kind of techno-skeptic.
That you know of. Every smartphone has a microphone, as do most laptops and desktops.
We still use wireless/laptop ofc and when we are among ourselves we have a rule that the phone goes in the box from 18:00 - 06:00. but for visitors there is no cell-phone in my house (this is perfect for us because we can pretty much rule out microphones/cameras being used).
I understand this might sound very strange for most people here, and it is strange. But how would you feel if say in the 80ies and 90ies you invite somebody for dinner and they show up with a big camera/flash around their neck and a microphone and tape recorder.
bienvenue chez moi!
also my partner does use a mobile phone for work, or on the go. I don't own one.
Consider other things that have a similar cost/benefit ratio that are more obviously mad:
* I could install a 3 metre high wall with razor wire around my house in case the police try to attack me for some reason.
* I could walk everywhere wearing a bicycle helmet in case I fall over.
* I could never use stairs. Only use lifts - they're much safer!
Hopefully it's obvious why those are mad.
I'm really conflicted, as your approach is privacy focused but I just can't imagine following it in real life. Perhaps I'm part of the problem...
If you ask their wearer to put them away before entering, then you can't have conversation with him/her.
you misunderstood. privacy is the conversation opener but discussions around mental health are the goal. when you say it is unhealthy are you implying "devices on the dinner table are healthy?" or putting toddlers on a screen/ipad is healthy? or showing your kids that it's ok to be only half present when interacting with them is healthy? sorry I utterly missed your point.
Seems more like daily Sabbath.
Many privacy-concious people opt for software that allows them to verifiably control their hardware, be it desktop/laptop or smartphone.
Some will go the extra step to physically remove or disable microphones and speakers.
I know a fair number of privacy-concious people, and not one of them have any sort of verifiable control of their hardware. It's surprisingly hard to do this, it turns out…
I have a reasonable level of confidence that my use of wirecutters and/or soldering iron gives me enough control over this kind of thing. It's impractical on a smartphone, but I solve that by not owning one.
Also houses have dogs or kids. Or both. It’s a thing.
Seems odd that there are older people who can hear the high-pitches when younger people can't. Maybe the kids are tuning it out.
There was a story a few years back about stores getting rid of young loiterers by playing these tones, then later on kids setting ringtones or alerts on their phones so that the teacher wouldn’t get mad that their phones were on. Guess what? A lot of surprised 15 year olds had their phones confiscated by disgruntled 40 year old teachers who could damn well hear the ringtones.
I had a "I haven't heard that in ages" moment over the summer; was in some shop in the UK that had a CRT showing their CCTV cameras and could hear the thing well before I could see it. Apparently my high-end hearing's still decent.
Due to the shortage, a "crt vault" was created. [0]
I remember visiting multiple thrift stores ~2 years ago to collect them. The most I had was 6 I think, but I ended up giving them to other players. Here's an example of what they look like at tournaments [1]
[0]: https://old.reddit.com/r/smashbros/comments/3dzqd9/to_those_...
I'm 39.
I was 35 at the time and my friends were 30ish, I guess it varies by person.
Does this include smartphones, computers, or other such devices capable of becoming always-on microphones in software?
I believe they were taking advantage of the non linearity of microphones.
Maybe I'll find some time to setup a self-hosted voice assistant. Recommendations?
Pretty sure you can hook it up to Mozilla DeepSpeech for completely offline stuff.
It seems like you have to register to create an account?
There is also a ‘skills’ marketplace, so not sure how private it is.
You do need to pair it with an account to activate basic skills, but:
> By default, Mycroft will not store your usage data or keep any recordings of your voice
https://mycroft-ai.gitbook.io/docs/using-mycroft-ai/pairing-...
"It is a bit of a chicken-and-the-egg situation, but I don’t think there is a strong case for creating an easy offline setup before the core components are ready. It is close, but I suspect it will be at least another year before we (the open source community) have all those pieces ready."
However, their backend seems to be on GitHub.
Yeah but not on a raspberry pi...which is what mycroft usually ends up on
Plus I'd say they chose by far the easiest target. A squirrel's sneezing is enough to confuse Alexa beyond recognition.
In my personal experience, the Google and Baidu speech to text algorithms are much more resilient against background noise.
[0]https://arstechnica.com/tech-policy/2015/11/beware-of-ads-th...
Although it does require knowing what the underlying natural sound to be modulated is (some processing required).
Robustness to background noise doesn’t necessarily help either.
Try adding a loud ultrasonic sine wave to speech in your favorite audio editor, and then hard clip it. Most of the distortion can be filtered out without too much damage to the speech by looking at the spectrogram and notch filtering unwanted frequencies. I imagine modern AI could do the same thing automatically.
But high-pass filtered white noise + clipping will make the speech unrecoverable if it's loud enough.
Turns into more cat and mouse at the end of the day.
That seems pretty cruel
It would also not be so hard to make such devices more privacy-friendly e.g. by giving users are clearer indication of when they're listening to you and when they transfers data to the cloud. I think Amazon already allows you to review and download all audio conversations they have from you (in theory this is mandatory in the EU).
Do people realistically think attackers are going to start burning zero days against these things to listen to the average Joe's kitchen conversations or bedroom chatter? What is the threat model here?
* The device is hardwired to light up whenever it is listening so you can't be eavesdropped on without some indicator.
* The device only transmits when it is in listening mode. This has been validated independently by multiple people.
* The device has a mute function (granted, software).
* The device can be registered under a bogus account.
* You can turn off all phone call capabilities / external listening capabilities.
* You can pihole it so it won't reach back to Amazon's ad services.
* You can delete any conversation you don't want stored manually, including one click deleting your entire history.
* You can have all content automatically deleted after X days.
So what exactly are people concerned about? Out of all the convos someone could have, a negative one just happens to probabilistically be recorded, you don't notice, and within 90 days someone in law enforcement or entity X figures it out and uses it against you? Someone at Amazon decides to look through your stored conversations and sees you turn on the lights at 5 PM each day?
I'm not saying there are 0 privacy implications with this device, as any device that has a mic and can record remotely has privacy implications. I just feel like the level of angst generated by smart speakers isn't warranted in the grander scheme of things.
Privacy is a niche topic right now and it might stay that way, but I think more and more people will become concerned about what large companies know about them. Each individual service and datapoint is not very critical as you say, combining data from multiple sources and using the data in unexpected ways can quickly reveal very intimate things about a person though. Companies like Amazon, Google and FB do not mainly use your data to “improve your experience”, they use it to nudge you into behavioral patterns that benefit their business model (i.e. look at more ads or buy more stuff). I doesn’t take so much imagination to see how this might harm you e.g. because you develop unhealthy spending habits or waste large amounts of time on social media. And the more data a company has about you, the easier it is for them to manipulate you. So for me it’s not about defending against an unknown attacker, it’s defending against companies using my data to manipulate me, which they might do even if they keep my data very, very secure.
We’re reaching the point where it becomes feasible in terms of electronics density and battery efficiency.
It does reignite the debate over what is acceptable. Cost of putting microphone in anything is negligible. Connection to internet in many things is assumed so even if microphone is unused, it could be turned on with an update.
Then again, my social circle met yesterday and they have multiple devices controlled by voice. It didn't stop anyone from saying things that would not look good if broadcast in public. It is weird, how easy it is to forget about it. Probably because you think it is your space. I would not dream of messing with their devices. I am a guest. I opted to be there.
I tried to think about it rationally. I can't give them too much shit. Between me and my wife, we have 8 devices in house now that have microphones and/or cameras in.
Ok google is off. Siri is off on phone, but on watch it is activated by a button, so there is semblance of control.
Do you have a source for that blanket statement?
Maybe they trying to pitch it as a fashion thing - like a social/personal statement. All these mitigations seems strange to me since mitigations are always an attack surface too.