• DarkCloud@lemmy.world
    link
    fedilink
    arrow-up
    74
    arrow-down
    1
    ·
    22 hours ago

    Side opinion: Both commenters are AI.

    This is a method of engagement farming. Any traffic is good traffic.

      • BenjiRenji@feddit.org
        link
        fedilink
        arrow-up
        13
        ·
        13 hours ago

        It actually doesn’t matter. Someone spent money to generate the content, someone spent money to run the bots. Money is being spent, the “economy” is bustling, nothing of value was created.

    • can@sh.itjust.worksOP
      link
      fedilink
      arrow-up
      2
      arrow-down
      1
      ·
      19 hours ago

      I would like to agree with you, but many of these commenters have innocuous accounts 4+ years old. Are they all compromised?

      • Hoimo@ani.social
        link
        fedilink
        arrow-up
        5
        ·
        12 hours ago

        4 is nothing, bot accounts were a massive industry even 10 years ago. They’d sell them to astroturfers and spammers to evade the most basic spam filters. Member the 2016 election? The years leading up to that, bot activity was noticably high. Then in the months before, suddenly every “black single mother” and “rural swing voter” had a 3 year old account and a boner for an orange maniac.

      • DarkCloud@lemmy.world
        link
        fedilink
        arrow-up
        6
        arrow-down
        1
        ·
        edit-2
        13 hours ago

        I’m not trying to get you to agree with me. I don’t care what you think about it. But content farms have obviously been around for various amounts of time (some older than four years, some newer)… And no, obviously not all accounts across all services are bots.

        You don’t seem to have put much thought into your questions, if you had you probably wouldn’t have asked them. If you just take a minute they answer themselves.

  • joeschmoe@lemmy.world
    link
    fedilink
    arrow-up
    40
    ·
    22 hours ago

    Just looking at the image, it’s clear that it’s AI. I mean, she’s made of fucking glossy plastic, lol.

  • Dejected Warp Core@lemmy.world
    link
    fedilink
    arrow-up
    36
    ·
    1 day ago

    It took me a minute. Then I noticed:

    • The guitar strap that attaches to the middle of the neck
    • The mic stand that just isn’t right, and the mic itself is just a weird size
    • The P E R F E C T glossy skin
    • If I could be bothered, I bet I could figure out that the fretboard is just fucked
    • Hair just looks plain airbrushed

    What concerns me is that anyone just on a rapid scroll-by might not notice any of that.

    • wittymcusername@lemmy.zip
      link
      fedilink
      arrow-up
      25
      ·
      23 hours ago

      The guitar strap that attaches to the middle of the neck

      Even as a guitar player, I was staring at her neck wondering wtf you were talking about for a few moments.

    • Omgpwnies@lemmy.world
      link
      fedilink
      English
      arrow-up
      5
      ·
      22 hours ago

      There’s not really and hard-and-fast rule about fretboard markers, but traditionally, markers are on odd frets from 3-9, then it skips two to mark 12 because that’s the mid-point of the string so the string’s root, 12th fret and 12th fret harmonic are the same note in different octaves. From there, it skips 2 again to the 15th fret and goes back to marking every other fret.

      Since the only gap of 2 that is visible is at the far-right hand side of the picture, you can assume that’s the 12th fret, meaning that a) that acoustic guitar has ~27 frets when they usually have ~20-ish and that the neck is probably a foot longer than it should be.

      Now, where the strap tries to attach to the neck is roughly correct if the guitar body was also over there since a standard-length guitar’s neck meets the body at about the 15th-17th fret. If you take my interpretation of the fret markers, that would put the strap at about the 15th fret.

      Going by what’s hanging on my wall, that would put the neck and scale length to be about similar to a bass guitar, but on a 6-string acoustic.

  • brucethemoose@lemmy.world
    link
    fedilink
    arrow-up
    113
    ·
    edit-2
    2 days ago

    I saw this Discord that was basically dedicated to AI influencer setups. People run these absolutely massive bot farms with tons of accounts, and do the whole stack. Baiting with photos, text chat, OnlyFans feeds…

    It was eye opening.

    It was interesting because it was just random individuals doing it. Some ran stuff on personal hardware or ex-crypto mining setups, some used various APIs.

    And it’s all so dystopian it’s… comical? They don’t make that much money individually, though Instagram is likely raking it in (and apparently cannot be bothered to police them; that’s not even a concern in the Discord).

    • Platforms exploit consumers by keeping them glued to their screen for hours at a time. They abuse producers by tricking them into slaving away making content, thinking they can become obscenely wealthy with passive income for little effort. The producers are themselves complicit in the abuse of the consumers, producing addictive content for a captive audience with lots of filler. The advertisers are also complicit. It’s a whole rotten show of exploitation. Some of your favorite creators have altered their content in one way or another that it makes them more money at the expense of your time or ability to focus. AI slop is the current stage, but it was rotten before that. Gig economy shit. Feudalism was better than this.

    • Rothe@piefed.social
      link
      fedilink
      English
      arrow-up
      17
      ·
      1 day ago

      LLMs has made it infinitely easier to be a scammer. Probably one of its top success stories. Mass surveillance is another thing it is relatively good at.

      And then the sloppy code of course.

      Lots and lots of benefits from the hundreds of billions of dollars thrown at it.

        • luciferofastora@feddit.org
          link
          fedilink
          arrow-up
          2
          ·
          15 hours ago

          …the ones that are actually built or in construction, anyway. Can’t really destroy announced 2024 to open in 2025 and still barely more than a clearing at the end of 2026. It seems like their construction schedules were drafted according to the “concept of a plan” principle.

      • brucethemoose@lemmy.world
        link
        fedilink
        arrow-up
        3
        ·
        1 day ago

        Eh, I think the bigger takeaway is how messed up Instagram and similar platforms are.

        The whole business model is just icky with real people twerking; replacing the performer simply makes it easier to see.

    • Jiral@lemmy.world
      link
      fedilink
      arrow-up
      16
      ·
      1 day ago

      That is a bigger issue for LLMs than many realise. It means that the models are fed increasingly their own excrements during training.

      • pebbles@sh.itjust.works
        cake
        link
        fedilink
        English
        arrow-up
        9
        ·
        24 hours ago

        The days of just scraping the internet are over. Data quality is a huge part of LLM training, I don’t think insta getting sloppified will save us unfortunately.

        • Jiral@lemmy.world
          link
          fedilink
          arrow-up
          1
          ·
          22 hours ago

          Yeah, but you need huge amounts of data to train, where to take it if not from the internet? I mean the existing books, records etc before the slop age are soon all fed as well.

          • pebbles@sh.itjust.works
            cake
            link
            fedilink
            English
            arrow-up
            1
            ·
            7 hours ago

            From my understanding a lot of books are still getting digitized, so there is a constant trickle of old literature and news papers and such. But really the answer is synthetic data. Data made by AI and other means for the purpose of training. Which that is very different than just insta slop.

            • Jiral@lemmy.world
              link
              fedilink
              arrow-up
              1
              ·
              3 hours ago

              The digitisation is done at such verociuous speed that this will soon slow down to a trickle because anythimg left is hard to come by.

              “Synthetic data” created with contribution from LLMs is poisining the models just as well. It is leading to reinforcement of hallucinations and artefacts.

              • pebbles@sh.itjust.works
                cake
                link
                fedilink
                English
                arrow-up
                1
                ·
                2 hours ago

                I would reccomend you look into sythetic data. It is a real thing and is used intentionally in basically every LLM of the past year or more. Last I heard of an LLM without sythetic data was Dots LLM1 and that wasn’t competitive when it came out.

          • luciferofastora@feddit.org
            link
            fedilink
            arrow-up
            2
            ·
            15 hours ago

            What they meant is that you could no longer just scrape the internet, but would need to filter the data for quality before mixing it into the slop substrate.

    • can@sh.itjust.worksOP
      link
      fedilink
      arrow-up
      3
      ·
      20 hours ago

      From why I hear from my partner it absolutely is. This screenshot is from YouTube however.