The Internet being mostly broken at this point is driving me a little insane, and I can’t believe that people who have the power to keep a functioning search engine for themselves wouldn’t go ahead and do it.

I wonder about this every time I see people(?) crowing about how amazing AI is. Like, is there some secret useful AI out there that plebs like me don’t get to use? Because otherwise, huh?

  • kromem@lemmy.world
    link
    fedilink
    English
    arrow-up
    3
    ·
    16 hours ago

    Gemini 3 Pro is pretty nuts already.

    But yes, labs have unreleased higher cost models. Like the OpenAI model that was thousands of dollars per ARC-AGI answer. Or limited release models with different post-training like the Claude for the DoD.

    When you talk about a secret useful AI — what are you trying to use AI for that you are feeling modern models are deficient in?

    • salacious_coaster@feddit.onlineOP
      link
      fedilink
      English
      arrow-up
      8
      ·
      16 hours ago

      what are you trying to use AI for that you are feeling modern models are deficient in?

      Giving correct answers more than half the time.

      • kromem@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        arrow-down
        1
        ·
        15 hours ago

        Can you give an example of a question where you feel like the answer is only correct half the time or less?

        • GrapheneOSRuinedMyPixel@sh.itjust.works
          link
          fedilink
          English
          arrow-up
          3
          ·
          14 hours ago

          steam deck as bluetooth controller:

          You can use the Steam Deck as a Bluetooth controller by enabling Bluetooth in the settings and pairing it with your PC or other devices. This allows you to utilize its controls for gaming on compatible platforms.

          • kromem@lemmy.world
            link
            fedilink
            English
            arrow-up
            2
            arrow-down
            2
            ·
            13 hours ago

            For future reference, when you ask questions about how to do something, it’s usually a good idea to also ask if the thing is possible.

            While models can do more than just extending the context, there still is a gravity to continuation.

            A good example of this would be if you ask what the seahorse emoji is. Because the phrasing suggests there is one, many models go in a loop trying to identify what it is. If instead you ask “is there a seahorse emoji and if so what is it” you’ll get them much more often landing on there not being the emoji as it’s introduced into the context’s consideration.

            • sem@piefed.blahaj.zone
              link
              fedilink
              English
              arrow-up
              3
              ·
              11 hours ago

              It is also not good at finding songs where you only (mis)remember some lyrics, but it will confidently invent a fake song.

              It also makes up citations.

              I wish it was better at expressing how likely its answer is to be true.

              • kromem@lemmy.world
                link
                fedilink
                English
                arrow-up
                2
                ·
                8 hours ago

                Yeah. The confabulation/hallucination thing is a real issue.

                OpenAI had some good research a few months ago that laid a lot of the blame on reinforcement learning that only rewards having the right answer vs correctly saying “I don’t know.” So they’re basically trained like taking tests where it’s always better to guess the answer than not provide an answer.

                But this leads to being full of shit when not knowing an answer or being more likely to make up an answer than say there isn’t one when what’s being asked is impossible.