• pipe01@programming.dev
    link
    fedilink
    arrow-up
    41
    arrow-down
    2
    ·
    7 days ago

    And surely the AI that companies control will never have any bias or misrepresent facts to fuel a narrative when used by people that don’t know any better because they have relegated all their thinking to a machine!

    • realitista@lemmus.org
      link
      fedilink
      English
      arrow-up
      6
      arrow-down
      42
      ·
      7 days ago

      This will definitely happen. But the current state of things is that AI’s are far more honest than right wing media for example. And even with Elon trying super hard to make Grok a bigoted right wing AI, it usually doesn’t toe the line and tells the truth instead.

      • athatet@lemmy.zip
        link
        fedilink
        arrow-up
        10
        arrow-down
        1
        ·
        7 days ago

        LLMs don’t tell the truth. They just string words together that would likely go next to each other.

        • realitista@lemmus.org
          link
          fedilink
          English
          arrow-up
          1
          arrow-down
          8
          ·
          7 days ago

          This is a stupid cop out. You can read something that an AI engine spits out and judge whether it’s true or not. And even on a technical level, modern AI engines do a lot more than just what we traditionally think of as an LLM do. They conduct research, gather data, transform it, process it and return results based on that. I mean, I told one to take a handwritten table, tansform it into an excel sheet and give it back to me. It did it more or less perfectly. How can that possibly be construed as just guessing the next word?

      • thedirtyknapkin@lemmy.world
        link
        fedilink
        arrow-up
        2
        ·
        edit-2
        6 days ago

        every currently problematic technology started as a honey pot that could be turned into a nightmare. then when they killed all the competition got turned it into a nightmare. we can’t ignore the nightmare that it will definitely become. regardless of what it’s luring people in with now. especially when you consider the men in charge of these things.

        we can’t keep falling for it. this one has the biggest implications yet. when they get good at using it to manipulate consumer behavior we’re absolutely fucked if the bottom half of the population is addicted. look at how easily they manipulate them anyway through social media. half of the pro trunp tiktoks in the last us election were just sad war clips with a caption like “me and the boys after kamala starts a war with iran”. it worked. now imagine if the thing that they let think for them tells them that trump’s opponent is bad and dumb every time they ask. elon is already openly trying to manipulate grok’s responses.

        but it’s impossible to say it better than cory doctorow’s original enshitification article. read that for more.

        ultimately the real issue isn’t how accurate they are. it’s giving your mind over to corporations that have a fiduciary duty to maximize profits for their shareholders