• ByteJunk@lemmy.world
      link
      fedilink
      English
      arrow-up
      4
      arrow-down
      3
      ·
      2 days ago

      ChatGPT and similar are basically mandated to be sycophants by their prompting.

      Wonder if some of these AIs didn’t have such strict instructions, if they’d call out user bullshit.

      • anomnom@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        7
        arrow-down
        1
        ·
        1 day ago

        Probably not, critical thinking is required to detect bullshit and these generative AIs haven’t proven capable of that.

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

          Fair point, but I’ll raise the counter argument that they were trained with a lot of internet data, where people slapping each other is the norm, and that seems suspiciously absent from AI interactions…