The Sapienza computer scientists say Wi-Fi signals offer superior surveillance potential compared to cameras because they’re not affected by light conditions, can penetrate walls and other obstacles, and they’re more privacy-preserving than visual images.

[…] The Rome-based researchers who proposed WhoFi claim their technique makes accurate matches on the public NTU-Fi dataset up to 95.5 percent of the time when the deep neural network uses the transformer encoding architecture.

  • besselj@lemmy.ca
    link
    fedilink
    English
    arrow-up
    157
    arrow-down
    1
    ·
    1 month ago

    Ironically, a tin foil hat would probably work to prevent that kind of surveillance

    • hornedfiend@sopuli.xyz
      link
      fedilink
      English
      arrow-up
      25
      ·
      edit-2
      1 month ago

      wouldn’t that make it worse? basically any signal can bounce off you, making yourself even easier to track.

      edit: wording

      • besselj@lemmy.ca
        link
        fedilink
        English
        arrow-up
        37
        ·
        1 month ago

        The tracking happens even with a big reflector/scatterer on your head, but as long as you dont wear it regularly, the system would have difficulty identifying you from wave propagation alone

      • Capricorn_Geriatric@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 month ago

        Since it ‘figerprints’ you, changing your fingerprint by blocking parts of the signal with pieces of foil doesn’t seem like a terrible idea.

        Now, the question is: is such a tactic like wearing gloves, or like using super glue?

  • hisao@ani.social
    link
    fedilink
    English
    arrow-up
    75
    arrow-down
    1
    ·
    1 month ago

    I’ve seen some article recently that the patterns of Wi-Fi/Bluetooth (don’t remember which one) interference with brainwaves can be scanned to reconstruct brainwave signature remotely, meaning that it might be possible to scan anyone’s EEG from Wi-Fi/Bluetooth distance. And there are some AI advancements for reconstructing inner monologue from EEG. So maybe we’re not so far from actual remote mind-reading.

  • Seleni@lemmy.world
    link
    fedilink
    English
    arrow-up
    66
    ·
    edit-2
    1 month ago

    accurate matches up to 95.5% of the time

    and they’re more privacy-preserving than visual images

    Oh fuck all the way off.

    • D_C@sh.itjust.works
      link
      fedilink
      English
      arrow-up
      10
      ·
      1 month ago

      When anyone or anything says that their product works “up to x%” I always presume it doesn’t really work at all.
      Christ, 1% is included in that “up to 95.5%” vague bullshit statement.

      • novus_dervish@lemmynsfw.com
        link
        fedilink
        English
        arrow-up
        2
        ·
        1 month ago

        I believe the reason they had to say “up to” is because the “signatute” will vary day to day ever so slightly (natural weight fluctuation), and if you gain or lose weight it can change dramatically, so the AI would have to constantly consider that and adjust it’s records.

        Honestly, unpopular opinion, but as long as it isn’t very short wavelength RF and they allow for self-hosted/open-source alternatives, I do find it a bit more privacy respecting than cameras, of course they have to say they are using the technology in public places.

        It also has it’s ways of fooling it, instead of wearing a wig and a false nose, you could wear a carbon-infused silicone fat suit to change the way you interact with RF.

      • toynbee@lemmy.world
        link
        fedilink
        English
        arrow-up
        2
        ·
        edit-2
        1 month ago

        I hate it when commercials say “up to 100%.” It’s literally a pointless metric; that could mean anything from 0% to 100%, inclusive.

        edit: Closed quote.

    • StenSaksTapir@feddit.dk
      link
      fedilink
      English
      arrow-up
      28
      ·
      1 month ago

      Well I heard about this and thought “this will be great for home automation”, but I also know that someone was equally excited about using this to rob people of basic freedoms or being a fucking creep or both.

      • Echo Dot@feddit.uk
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 month ago

        If it’s your home why can’t you just have a camera or motion sensor. Rather than trying to adapt something that isn’t designed for the purpose.

        • StenSaksTapir@feddit.dk
          link
          fedilink
          English
          arrow-up
          6
          ·
          1 month ago

          Cameras require light, while radio waves works almost as well in darkness.

          A motion sensor is an extra device that needs to be connected, have power and so on.

          There are already radio wave motion- and room occupancy sensors where you can specify zones and so on, but if I could have personalized on top of that I’d take it.

          Finally, using a thing for something useful other than its intended purpose is kinda fun.

    • gcheliotis@lemmy.world
      link
      fedilink
      English
      arrow-up
      25
      ·
      1 month ago

      I can imagine this being initially an accidental discovery like oh every time so and so’s body interacts with the WiFi signal it’s the same pattern… until someone starts exploring this further… and then some engineer or their manager started looking for applications for this. In my experience engineering researchers especially are very good with coming up with use cases for whatever tech they’re working with, with little ethical consideration.

      • sugar_in_your_tea@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        7
        ·
        1 month ago

        I doubt it. You’d need to be looking really closely at the waveforms to notice this, so they were likely already doing something similar, like that research that can pinpoint where people are in a house based on their WiFi. They were probably already doing something creepy before they noticed that this was more straightforward than they expected.

        • turtlesareneat@discuss.online
          link
          fedilink
          English
          arrow-up
          4
          ·
          1 month ago

          Once you start playing with radiowaves and antenna you start noticing the intricate ways it plays with and around bags of water like bodies. I’m sure the original research on location/movement tracking was due to scientists trying not to get interference, later once they figured it out it was natural to see how much data they could get out of a radio interference profile.

          I remember the original tech was going to be marketed as a way to tell if your old person (parent etc) had fallen down and stopped moving. Not the best use case, and then the privacy implications became clear. Once that happens the race begins to exploit the tech.

          …But the eventuality here is something like a Star Trek tricorder that can take multiple vitals and detect irregularities from across the waiting room. Sensors that remember who was in a room and what settings they had. Etc. Some cool thing besides the bad stuff (microtarget those ads).

    • douglasg14b@lemmy.world
      link
      fedilink
      English
      arrow-up
      15
      ·
      1 month ago

      Everything is incremental progress in some way.

      I remember years back someone doing experiments with Wi-Fi to see if a room was occupied based on signal attenuation.

      This just looks like an extension of that.

      Not everything is a giant leap

    • MouldyCat@feddit.uk
      link
      fedilink
      English
      arrow-up
      11
      ·
      1 month ago

      You think if people who publish their work publicly didn’t research things like this, they would just never be discovered?

      At least this way, we all know about the possibility, and further research can be done to see what can mitigate it.

  • Sabata@ani.social
    link
    fedilink
    English
    arrow-up
    52
    ·
    1 month ago

    Incorrect bio-signature detected, drink verification can to continue your content.

  • artyom@piefed.social
    link
    fedilink
    English
    arrow-up
    39
    ·
    1 month ago

    they’re more privacy-preserving than visual images.

    hhhhwat. How can they identify you and also be privacy preserving? 🤔

    • sorter_plainview@lemmy.today
      link
      fedilink
      English
      arrow-up
      19
      arrow-down
      1
      ·
      1 month ago

      It’s all AI. You should not worry about it. In fact you should not think about it. All is going to be fine.

    • Diplomjodler@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      ·
      1 month ago

      Well, the alternative would be a camera in every toilet stall. See how our benevolent corporate overlords only have our best interest in mind?

    • Vinstaal0@feddit.nl
      link
      fedilink
      English
      arrow-up
      2
      ·
      1 month ago

      They know you are a person and they can call your a certain UUID, but there will be a hard time matching you to your name etc.

      Camera’s can do face recognition (if your face is even in the database) to know who you are.

      This only works until the point where they have your form in a database which they can check…

      • artyom@piefed.social
        link
        fedilink
        English
        arrow-up
        5
        ·
        1 month ago

        We have heard this non-sense before, only to find it’s trivially easy to connect to your PID.

        • Vinstaal0@feddit.nl
          link
          fedilink
          English
          arrow-up
          1
          ·
          1 month ago

          Never said that it wasn’t easy, it’s just harder than with facial recognition. In theory you could do it correctly in a way that it isn’t indentifiable.

          Also this works in places where faces are protected

    • Empricorn@feddit.nl
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 month ago

      I’d imagine it’s like online advertisers: they convert your fingerprint to a token to try to sell you shit, but they allegedly don’t know who exactly you are or where you go. So visiting animatedllamaporn.com is still your little secret…

        • realitista@lemmus.org
          link
          fedilink
          English
          arrow-up
          2
          ·
          edit-2
          1 month ago

          Well they can identify you are the same person but not your identity… So it’s like a disenbodied fingerprint.

          I suppose they could potentially make some database and train an AI on it someday to match to actual identities, but usefulness would be pretty limited at only 95% accuracy. That’s a false reading 1/20 times, so I suspect it would fail bigly to accurately recognize people from large data sets.

  • LillyPip@lemmy.ca
    link
    fedilink
    English
    arrow-up
    27
    arrow-down
    1
    ·
    1 month ago

    Neat. Good luck protecting yourself from this.

    On the other hand, I’m seriously considering opening an Etsy shop selling foil-lined clothes. I’m pretty good at sewing. What do you think?

    • Krudler@lemmy.world
      link
      fedilink
      English
      arrow-up
      3
      ·
      edit-2
      1 month ago

      The most primitive of physics concepts, the transmission/absorption/reflection of energy, is completely unknown to most people it would seem.

    • Telorand@reddthat.com
      link
      fedilink
      English
      arrow-up
      10
      ·
      1 month ago

      Ironically, they’re still wrong, because even in their wildest conspiracies, they didn’t imagine Wi-Fi could be used to “take pictures” of a sort.