• @[email protected]
    link
    fedilink
    41 month ago

    I switched to Kagi in january 2022 and havent even tried Google search since then. So it has an Ai telling people to kill themselves now? Seems par for the course.

    • @[email protected]
      link
      fedilink
      English
      7
      edit-2
      1 month ago

      Kagi is incredible, my favorite feature is being able to apply my own weights to sites. Ad driven models wouldn’t allow this because ads are all about controlling what the user sees. I feel like I’m in control and it’s well worth the $10 to me

  • slazer2au
    link
    fedilink
    English
    11 month ago

    Na, your the moron here. How do you expect the page to know how many ports you have without giving it your mobo make and model.

    • @[email protected]
      link
      fedilink
      11 month ago

      Why are you talking so mean to sodium? Sodium probably didn’t ask the question! It’s just an element!

      Also, *you’re and *?.

    • @[email protected]
      link
      fedilink
      21 month ago

      Not sure if this is a joke, but in case it is not: the relevant part of the post is not the highlighted sentence, but what comes right after that.

      • Ekky
        link
        fedilink
        English
        51 month ago

        This might have been one of those cases where a overly obvious red arrow and obnoxious circles might be needed.

        • @[email protected]
          link
          fedilink
          1
          edit-2
          1 month ago

          I’d go the other way… if there was no highlight, I bet people would get it, but they’re focusing on the part that’s explicitly called out.

  • @[email protected]
    link
    fedilink
    9
    edit-2
    1 month ago

    Hey some tankie just sent me the same thing on Lemmy, this really is the next reddit! 10/10 advice

    • @[email protected]
      link
      fedilink
      01 month ago

      It’s universal when 1 in 100 anonymous people are in fact chaotic assholes (and the rest, lerkers). Just a bad idea to train AI on the whole of us and not just the best of us.

      • @[email protected]
        link
        fedilink
        11 month ago

        oh it doesn’t bother me. I hit him with the ‘no u’ and went on with my day. The AI’s seem to be pretty based so far. I think thats why they are rapidly trying to remove real accounts and flood reddit with bot accounts. The only way they can manipulate the AI is to manipulate the data. Thus far its been good at generating information for me, and seems to have some enlightened and woke takes. I’ve mostly been using Grok and I can see why having another one of his tube babies reject him is hard.

        • @[email protected]
          link
          fedilink
          English
          01 month ago

          Think of how many hurt feelings we’d all have to live with if they hadn’t invented rubber and glue.

          • @[email protected]
            link
            fedilink
            11 month ago

            Ah yes us painters know well that painting is such a relieving art because… sniffing all the paint.

  • sheepishly
    link
    fedilink
    01 month ago

    That reddit user? Me (it’s not me this is joke don’t ban me)

  • @[email protected]
    link
    fedilink
    English
    121 month ago

    Showing, once again, how wrong these “AI-assisted” results can be: it was definitely more than one user.

  • @[email protected]
    link
    fedilink
    English
    1
    edit-2
    1 month ago

    Google and Reddit are both 💩 so no surprise it gives that type of response from their 💩 ai.

  • @[email protected]
    link
    fedilink
    10
    edit-2
    1 month ago

    I bet it started with a genuine answer then got a lot of upvotes then original user decided to change the answer, probably due to reddit exodus thing.