I changed the title from “Spying” to “Eavesdropping” because the article actually directly supports that it is “spying” on you, just not listening.

  • EmptySlime@lemmy.blahaj.zone
    link
    fedilink
    English
    arrow-up
    1
    ·
    2 days ago

    I think I get what it’s saying? It’s saying that while your phone isn’t directly listening to your conversations in any meaningful way they collect crazy amounts of other data on basically everyone and can piece it together in such a way that they can make some scary accurate guesses as to the kind of ads to serve you based on what their systems have gathered your interests are and where/with whom you spend your time.

    I’m not entirely sure. They didn’t really seem to present much more than speculation on it.

    • Jack_Burton@lemmy.world
      link
      fedilink
      English
      arrow-up
      1
      ·
      2 days ago

      A while back on Reddit I saw a post asking about this stuff. Companies don’t need to “listen” anymore, they have much more sophisticated options now. This example will use 3 people: A (wife) B (husband) and C (wife’s old friend from school).

      The question: A goes to the store without B, and runs into C, who proceeds to tell A about this cool gaming chair he just got. After the conversation, A puts the interaction aside and never mentions it to B. B later gets ads for the gaming chair. If B never had any interaction whatsoever about the chair, and A never even talked about it to B, how does B get the ads?

      The answer: A goes to the store, and her phone knows this through location data. The algorithm knows A is at the store, and now picks up that C is also at the same store. The algo then finds a connection through social media that A and C know each other, and maybe even knows spending habits and sees A and C buy similar things. The odds are good that A and C will interact at the store.

      C has been searching about this gaming chair for months, has just recently bought it, and talks about it constantly on socials. Odds are good that if A and C interact, C will talk about the chair.

      A has no interest in gaming or tech, but B does. The algo knows A and B are married, and B would be interested in the chair C just bought. There is now a vector to send ads from the interaction of A and C directly to B, even though A never mentioned anything about the chair to B, and B has never even met C.

    • deranger@sh.itjust.works
      link
      fedilink
      English
      arrow-up
      2
      ·
      2 days ago

      This is what I’ve been saying for years. You don’t need to listen to someone’s microphone to serve eerily relevant ads. I’ve heard people commonly discussing how they talked about something and saw an ad for it later. You’re already being tracked everywhere and a bit of confirmation bias is all you need to focus in on the times it works. It’s like that story of the prenatal vitamins being recommended to that woman who didn’t realize she’s pregnant.

      This isn’t to say that I don’t believe someone can’t possibly turn on the mic in a targeted attack, but few of us are having conversations that are that important. It’s way easier to target you other ways using data that’s much more available.

      • BeMoreCareful@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 day ago

        I’ve been thinking it’s the other way around. You see such and such ad X times and then the next thing you know you’re thinking about it, then mention something to someone. Then Notice the ad you’ve been seeing for a while now.

        They don’t have to listen for a thought they put there in the first place.

        I think history will look back at this period of wild ass mass propaganda and be like: what do you mean they used it to sell crap?

    • Pheonixdown@lemm.ee
      link
      fedilink
      English
      arrow-up
      0
      ·
      2 days ago

      See, people say that ad companies can use all this information they gather to better serve targeted advertising, but that’s just not my anecdotal experience.

      I get served ads all the time in languages I don’t speak, for VERY specific job related audiences that I’m not even close to related to, state politics that I’ve never lived in, services that I’m already actively subscribed to, just the worst targeting ever.

      If I have to get advertised to, I’d so much rather get an ad that could actually be at all relevant to my life, or even some generic ad over the total misses.

      Like, if you’re going through the trouble to do all this shady shit to get my data at least be good at it using it…

      • taladar@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        1
        ·
        2 days ago

        If targeted advertising worked the way they say it does then why is Amazon with my entire purchase history at their disposal, still unable to stop themselves from trying to sell me a second washing machine just after I bought one from them? Or Audible with hundreds of books in my account, most of them English, is still trying to sell me German versions of books with original English language versions? The whole notion that advertising has all that data to do better advertising assumes a competence level that just isn’t there.