Hey Purism can you fight Alexa? --> Device idea

Some of you might already have read this thing here.

It is about a latest dev of a bracelet disturbing the Alexas and Siris in our world. Purism is currently focusing on devices with built in security and privacy. I would be very curious, how you guys and especially officiallly Purism is thinking about developing and selling devices like this that purely aim for fighting the big tech companies and their spies.

@Kyle_Rankin, what do you think?

It’s not really in the scope of what purism sells. Though I would imagine (a bit of emphasis on the word “imagine”) that perhaps the phone could be programmed to emit the same frequencies from its microphone. Or is that not feasible with a phone mic?

The device was already discussed in a restricted category (registered users only, hence you can view it, but not publicly visible).

I would hope that beyond getting the Librem 5 out the door, Purism would be more interested in creating an open spyhome device rather than combating those same devices as produced by other manufacturers. However there are so many surveillance capitalism devices that could beneficially be supplanted by open alternatives that there may be many that Purism never gets to. See also Your Purism products wish list

You didn’t ask me but…
You could get a mycroft, or just use the clapper (takes about 30 seconds to reconnect to everything though) with alexa (whose development kit is MIT licensed and she can only keep her microphone open for 8 seconds btw).

I think it would be interesting to see if the Librem 5’s loudest speaker (the bottom speakerphone one) is capable of producing those same high-frequency tones, loud enough, to perform the same function. If so it probably wouldn’t be too difficult to write up a simple app with yad that used a CLI audio playing tool to play those tones in a loop at max volume.


I couldn’t find anywhere where the text explicitly documented what frequency to use.

I’m thinking more like Rock’em Sock’em Robots:


And how long for Alexa to employ a low-pass filter? I guess one firmware update. :slight_smile:


I’m not sure they’d be willing to tackle the PR task of explaining why they’re defeating a mechanism intended to keep their device from spying on people.

1 Like

Do you think so? I think most people are sheep and wouldn’t care.

People like us who care don’t own them so it doesn’t matter if we kick off and the people who do own them clearly just don’t care that they’re being spied on.

There would an amount of kick back from the media, alt news and some publications such as the Guardian in the UK would kick off but in a weeks time it would be over, everyone would move on and forget that it ever happened.

How long would that take, though, for everyone to move on? Long enough to lose millions of dollars? I would think so.

It’s a bit of both.

It would be hard for Amazon PR to spin the surveillance. Some people would absorb the negativity about the product even if they didn’t know or understand exactly why it is a problem.


The media circus would move on. Amazon could calculate that the short term losses would be outweighed by the long term gains.

Regardless, Amazon would have to get caught countering the jamming. So it’s good that researchers are spending their time looking into what goes on.

I don’t see a pressing reason for Amazon to counter the jamming. If a person has clearly expressed a desire not to be recorded / listened in on, by wearing a jamming device, then countering the jamming serves what purpose? The unmodified device still works perfectly for those customers who choose to use the device, who choose not to jam, who choose to be recorded / listened in on.

Even an otherwise keen Alexa customer might want some means of jamming the device for periods of time e.g. while having sex (and, yes, the New York Times article notes that for quality assurance purposes LOL device makers have reviewed recordings of customers having sex).

Clearly though there is a difference between whether Amazon does counter the jamming (their choice) and whether Amazon can counter the jamming (too easy).



:slight_smile: is that something they could actually make money out of ? isn’t that just a repetitive string of silables and moans untill finally silence ?

You would have to ask the device makers.

1 Like

Couple of things on this one:

  1. emitting sound, regardless of the frequency, is a battery sink. Dedicated device seems to be the way to go if this is your cup.

  2. a LPF would be all that is needed, that and a little sound EQ with some additional gain, depending on how aggressive the cutoff on the filter is.

In theory this is a good idea, but it does seem like it can be easily defeated.

1 Like

That concerned me too.

Technologically speaking, yes, it could easily be defeated, but would it be worth it for them? I think maybe not. After all, it (alexa) has a hardware mute button. The only thing stopping people from keeping it muted unless they have a reason to talk to it is laziness/convenience. I believe if someone wrote an app that served as a remote mute activation app Amazon wouldn’t bother with trying to defeat that, and this is effectively the same thing.

1 Like

The Echo products do have a microphone mute button, and the ring around the top glows red when it is muted, but how do we know that it really is completely muted?

Furthermore, because it is rather trivial to defeat, I wouldn’t be so sure that they just wouldn’t give the Echo’s the ability to detect when they are being ‘jammed’ and just play along, while still recording the filtered stream.

However, I guess depending on the amplitude of the tone being emitted, it could crowd out others sounds that the microphone might have picked up normally. Amplitude is an equation based on power. So once again, how much juice for an effectively strong output of noise?

1 Like

They could just play along, or even lie about being truly muted, but they’ll get caught eventually and then there goes their credibility. Afterward, everyone will throw out their echos and get google home things. I still say it wouldn’t be worth the risk.

As for the power required to pull that off, I would imagine (not having an electrical background) that it would be “too much,” at least for “always on.” Phones play music for extended periods, though. Once in a while could be ok, perhaps.

I wonder how hard it would be for an android novice (like me) to create an android app to test a phone speaker would be? Certainly not that hard?

As there are android music players with open source available, it would be trivial at best, and just require a bit of your time.