Tech experts are starting to doubt that ChatGPT and A.I. ‘hallucinations’ will ever go away: ‘This isn’t fixable’::Experts are starting to doubt it, and even OpenAI CEO Sam Altman is a bit stumped.

  • Taringano
    link
    fedilink
    English
    arrow-up
    51
    arrow-down
    10
    ·
    1 year ago

    People make a big deal out of this but they forget humans will make shit up all the time.

    • Cybermass@lemmy.world
      link
      fedilink
      English
      arrow-up
      35
      arrow-down
      6
      ·
      1 year ago

      Yeah but humans can use critical thinking, even on themselves when they make shit up. I’ve definitely said something and then thought to myself “wait that doesn’t make sense for x reason, that can’t be right” and then I research and correct myself.

      AI is incapable of this.

      • bric
        link
        fedilink
        English
        arrow-up
        8
        arrow-down
        1
        ·
        1 year ago

        We think in multiple passes though, we have system 1 that thinks fast and makes mistakes, and we have a system 2 that works slower and thinks critically about the things going on in our brain, that’s how we correct ourselves. ChatGPT works a lot like our system 1, it goes with the most likely response without thinking, but there’s no reason that it can’t be one part of a multistep system that has self analysis like we do. It isn’t incapable of that, it just hasn’t been built yet

        • tweeks@feddit.nl
          link
          fedilink
          English
          arrow-up
          4
          arrow-down
          1
          ·
          1 year ago

          Exactly, if you replicate this behaviour with a “system 2 AI” correcting the main one it will probably give similar results as most of us.

          Heck you can eventually have 5 separate AIs discussing things out for you and then presenting the answer, at top speed.

          It will never be perfect, but it will outmatch humans soon enough.

      • Bitswap@lemmy.world
        link
        fedilink
        English
        arrow-up
        7
        arrow-down
        1
        ·
        1 year ago

        Can’t do this YET one method to reduce this could be to: create a response to query, then before responding to the human, check if answer is insane by querying a separate instance trained slightly differently…

        Give it time. We will get past this.

        • Cybermass@lemmy.world
          link
          fedilink
          English
          arrow-up
          3
          ·
          1 year ago

          We will need an entirely different type of AI that functions on an inherently different structure to get past this hurdle, but yes I do agree it will eventually happen.

          • Bitswap@lemmy.world
            link
            fedilink
            English
            arrow-up
            1
            ·
            1 year ago

            Agreed. This will not come from a LLM…but honestly don’t think it’s that far off.

      • Taringano
        link
        fedilink
        English
        arrow-up
        7
        arrow-down
        4
        ·
        1 year ago

        You’re just being victim of your own biases. You only notice that was the case when you were successful in Detecting your hallucinations. You wouldn’t know if you made stuff up by accident and nobody noticed, not even you.

        Whereas we are checking 100% of th AI responses, do we check 100% of our responses?

        Sure it’s not the same thing or AI might do more, but the problem is your example. Where people think they are infallible because of their biases. when it’s not the case at all. We are imperfect, and we overlook our shortcomings possibly foregoing a better solution because of this. Because we measure the AI objectively, but we don’t measure what we compare it to.

        • Cybermass@lemmy.world
          link
          fedilink
          English
          arrow-up
          4
          arrow-down
          2
          ·
          1 year ago

          I never said we always question ourselves I just said that AI can’t so your entire reply doesn’t apply here