141–150 of 734 posts
Earlier quoted context omitted.
Nevermind that we are already in a world where you can play any song you like from the wireless computer you carry in your pocket. That's not convenient enough. /s
That pocket computer also has all the same potential threats to privacy. Somehow nobody cares about them, though.
Earlier quoted context omitted.
That pocket computer also has all the same potential threats to privacy. Somehow nobody cares about them, though.
That pocket computer can be programed to listen, or not. The whole point of these cylinders is to be listening at all times.
My guess is they mostly do work that way but bugs and vulnerabilities don't seem like a stretch. (As is to be expected with all software)
Earlier quoted context omitted.
There is a massive difference between walking a few steps to turn on a light switch (which is only useful when you are in close proximity) vs having a accurate real-time mapping functionality when on the go. Similarly, there is a massive difference in the privacy implications of listening to every conversation everyone in your home (including guests) is having vs. having your current location known. Wiretapping laws…
Turning off all the light switches that have been left on in a big house though is a meaningful convenience. I think the privacy point is strong enough without having to trash people's use cases.
People are volunteerelly paying for their houses to be tapped for the convenience of being able to shout: play me some song. I’m so out of touch with this world it’s scary.
- people in the sixties: "the government will wiretap your home"
- people now: "hey wiretap, can cats eat pancakes?"
Earlier quoted context omitted.
Do you notice that it randomly fades out and then fades back in every few hours? I don't understand why it does that.
Is it pulling a track from their music archives? Maybe it’s just a several hour long loop and the date is when it repeats.
That's the "Facial recognition fooled by funny-colored glasses" study from Carngie Mellon, where researchers were able to make machine learning algorithms fail disastrously (e.g. mistake a man for Milla Jovovich) with a pair of glasses printed with what looks like a random assortment of colorful pixels, but is in fact a targeted attack specifically designed to trick the algorithm.
This Alexa failure is obviously not a targeted attack; but when your system is exposed to enough data, eventually you will stumble over some input that happens to resemble a targeted attack by pure chance, right? It's equivalent to saying that if you aimed the facial recognition algos in that paper at millions of faces wearing randomly-colored glasses, eventually some of the glasses would be close enough to the targeted-attack glasses to produce the same effect.
Obviously I'm theorizing on almost no data here - I don't know anything about Alexa's voice recognition, maybe it contains no ML at all. But it seems plausible that this might be what happened here - not a bug per se, but the natural and totally expected result of giving an opaque, machine-generated system with a very low failure rate so much input data that the failure rate is significant.
I feel creeped out by these home listening devices and I don't own one, but don't our phones already have this capability? You can turn "Ok Google" on on an android phone. I sometimes record audio, and the mic is incredibly good. Is there a substantial difference between our phones and these devices? EDIT: Just realized the substantial difference is that Google and Amazon own all of these things. They don't control a…
Not sure about Android phones with "Ok Google" but on iOS, all of Siri's voice processing is on the device. As opposed to Amazon Alexa which sends the data to the cloud for processing.
People are volunteerelly paying for their houses to be tapped for the convenience of being able to shout: play me some song. I’m so out of touch with this world it’s scary.
I like the idea, but wonder if the same features are possible on a stand alone tool. For example, the device does the NLP there on the premise. It exposes a plugin API for apps. The app gets the text translation by registering a callback URI. Then it can do want ever it wants with that. Expose a streaming receiver api for music. Essential get rid of the middle man service. Is this possible of a Pi?
I have this idea of a system I would like to have in my house. It contains cameras in every room that are constantly watching where people are and relaying the coordinates to a central server. That server makes decisions on if lights should be on or if A/C should be running in that room. But I would never buy this system. I would have to make it myself. I am hopeful that open source software and hardware can produce…
What actually happened: Alexa misinterpreted some voice commands and activated a "call" skill. The people involved and local news got very excited and escalated this into a conspiracy story. Amazon takes customer privacy EXTREMELY seriously. There's no way a team would get the "ok" to build a skill that randomly records private conversations then sends them to a random contact. It also doesn't make any logical sense…
If there was an Alexa in my place of residence, I would rip it out of the wall, smash it to bits with a hammer, and fire it out of a cannon into the sun.
Hell no to this crap from my side. I hope there are more incidents like this until people wake up to the fact they are putting telescreens in their homes. You guys can't be trusted!