• TrackinDaKraken@lemmy.world
    link
    fedilink
    English
    arrow-up
    65
    ·
    il y a 3 mois

    Yeah, I’m fine with giving up something I never used.

    And, from what I’ve read many of the people who use it for work only do so to appease their idiot bosses.

  • CanadaPlus@lemmy.sdf.org
    link
    fedilink
    arrow-up
    62
    ·
    il y a 3 mois

    Does this guy actually think the average AI-hater is also addicted to it?

    That person probably exists, but they’re nowhere close to the majority.

    • SayCyberOnceMore@feddit.uk
      link
      fedilink
      English
      arrow-up
      5
      ·
      il y a 3 mois

      I think the number of average, non-tech users is climbing rapidly… they just don’t realise how it’s enshitifying their lives yet.

      A bit like the <insert material> content in your <inseet posession>… are you aware of how bad that’s getting?

  • taiyang@lemmy.world
    link
    fedilink
    arrow-up
    42
    ·
    edit-2
    il y a 3 mois

    I admit, I’m tempted to read the AI answer when I search something; but holy crap is it a patronizing waste of time. Just now I was searching about a game and the long page of AI answer essentially boiled down to “kill stuff to get stronger,” like brah, I know, I’m trying to look up exact formulas here, you piece of shit.

    My questions are never answered by it, and when it tries, it’s wrong even with follow up info. Man I really could live without it, lol.

    Edit: oh, and this is a relatively new game but on both Google and Bing indexes, the first page and a half are clearly AI generated. Same boilerplate nothingburger answers. No surprise the search AI isn’t good, it’s regurgitating other AI answers. MAN I could live without it.

    • Malix@sopuli.xyz
      link
      fedilink
      arrow-up
      16
      ·
      edit-2
      il y a 3 mois

      The stupid thing about ai’s seems to be that they are so confidently wrong/guessing when they don’t have the slightest idea. Almost like they’ve been told to never say “I don’t know”.

      • zarkanian@sh.itjust.works
        link
        fedilink
        arrow-up
        1
        ·
        il y a 3 mois

        All they are doing is giving you a string of words that statistically follow each other. It can’t say “I don’t know” because it has no criteria for knowing.

    • FatVegan@leminal.space
      link
      fedilink
      arrow-up
      3
      ·
      il y a 3 mois

      I think the AI summary is one of the only vaiable things if it comes to AI. And i mean just to figure out if you wanna read into it, not even to use it in any other way.

  • joelfromaus@aussie.zone
    link
    fedilink
    English
    arrow-up
    28
    ·
    il y a 3 mois

    If you don’t use AI of course you won’t miss it.

    I use AI and still want it to fuck off.

    Spoiler

    It’s integrated into some systems I use at work and while it can be handy I wouldn’t give a shit if it was gone tomorrow since it’s better for humanity as a whole if it’s gone.

  • hperrin@lemmy.ca
    link
    fedilink
    English
    arrow-up
    25
    ·
    il y a 3 mois

    Tell me you’ve completely forgotten how to do your job without telling me you’ve completely forgotten how to do your job.

  • onnekas@sopuli.xyz
    link
    fedilink
    arrow-up
    22
    ·
    il y a 3 mois

    We got through workdays without LLMs only a FEW years ago. Except the hype for AI not much has changed since then. This alone should be proof that it’s possible.

    • vanillama@programming.dev
      link
      fedilink
      arrow-up
      10
      ·
      il y a 3 mois

      Yeah, if it weren’t for organizations adapting all their workflows and processes to involve as much LLM as possible it probably wouldn’t be as costly to eventually switch back to minimizing its usage to specific, more useful stuff (like debugging)

  • affenlehrer@feddit.org
    link
    fedilink
    arrow-up
    18
    ·
    il y a 3 mois

    I like LLMs even with their (big and inherent) flaws.

    However, I also don’t like it to be sprinkled everywhere. I want affordable smaller LLMs that run locally on my own hardware. RAM prices right now make that financially impossible for me.

    That way I’m the one who has to deal with the flaws (even more with smaller LLMs) and the energy prices but on the other hand it’s private and I have sovereignty. Nobody can turn off features or change the model or increase subscription prices or change the API or steal my data.

    • hperrin@lemmy.ca
      link
      fedilink
      English
      arrow-up
      10
      ·
      il y a 3 mois

      There are no stupid questi…

      Ok, there’s at least one stupid question.

  • Ethan@programming.dev
    link
    fedilink
    English
    arrow-up
    2
    ·
    il y a 3 mois

    If LLMs (since AI is not a thing that actually exists) were banned tomorrow, I’d be pretty grumpy that I had to go back to using the garbage fire that is modern search engines. And I’d be grumpy that I had to go back to writing boilerplate code myself. And then I’d get over it and move on with my life.