Incorrect AI-generated answers are forming a feedback loop of misinformation online.

  • Margot Robbie@lemmy.world
    link
    fedilink
    English
    arrow-up
    5
    arrow-down
    1
    ·
    1 year ago

    Who knew the Abominable Intelligence is actually really, really dumb.

    From my tests LLMs are only good at writing boring work emails, you should pretty much never trust any factual information they generate without verification.

    • Meowoem@sh.itjust.works
      link
      fedilink
      English
      arrow-up
      1
      arrow-down
      3
      ·
      1 year ago

      They’re amazing for so many things, coding and discovery especially - yes you need to verify answers but you were supposed to do that with Wikipedia and everything else anyway.