• CriticalResist8@lemmygrad.ml
    link
    fedilink
    arrow-up
    13
    ·
    11 days ago

    Not bad advice when it comes to how to organize around AI in the workplace, but by diminishing its capabilities (reducing it to ‘producing buggy code’ and ‘badly summarizing data’ at several points in the article), they will be making workers think that it’s not a big problem to think about, so it diminishes the message they want to convey. The reality is developers at all levels are already using agentic methods a lot. Can’t speak to numbers but it’s widespread, and they’re doing it themselves without management telling them to - so on that for example I agree with the article that workers should be able to choose how they integrate AI in their workspace, just like they should be able to choose their preferred tools and methods of work.

    • chobeat@lemmy.mlOP
      link
      fedilink
      arrow-up
      6
      arrow-down
      4
      ·
      11 days ago

      That’s contradictory with the direct experience of most workers, where AI is something forced upon them by managers and that creates more problem than it solves. The minority using agentic AI is also probably harder to organize, so better relate to the majority that is discontent than the minority that is vibe-coding.

      • ☆ Yσɠƚԋσʂ ☆@lemmygrad.ml
        link
        fedilink
        arrow-up
        7
        ·
        11 days ago

        I’ve been doing software dev for over two decades now. These tools absolutely do work, and they can save you a lot of time. The reality is that these tools are still very new, and people are learning how to use them effectively. They’re not magic, and you don’t just type a prompt and get a working program. You have to spend the time to actually learn how to use them effectively. Most people haven’t actually done that, especially the ones that complain about them most incessantly. I’ve seen this happen a lot personally. Where people don’t want these tools to work. They try them, and then when they don’t magically do what they want use that as proof that they don’t work.

        Basing the argument on these tools producing buggy code, not being effective, and not saving time, is just building a straw man. There are plenty of good arguments for organizing that are rooted in reality. There’s no need to invent fake arguments here.

        • burlemarx@lemmygrad.mlM
          link
          fedilink
          arrow-up
          1
          ·
          6 days ago

          Yes, I feel the same. While I do think those tools work, the productivity gains are not the end of the world, specially because if you want to use them effectively you need to be very specific and then review the generated code afterwards.

          The problem is not the AI tools per se, but the expectation that AI will be the new industrial revolution and the feeling that managers will be able to cut costs (and personnel) due to the fictional gains that AI provides. It ends up becoming a self-fufilling prophecy, since they force devs to work harder and longer with higher productivity targets.

          I do believe AI will be a big flop. Not because it is a bad tool, but because it will never fulfill the hype.

          • ☆ Yσɠƚԋσʂ ☆@lemmygrad.ml
            link
            fedilink
            arrow-up
            4
            ·
            6 days ago

            I’m kind of expecting we’ll see a similar scenario to the dotCom bubble which wiped out majority of the existing tech companies, but then once the dust settled, useful tech came out of the whole thing. We’re in the main hype phase of this technology right now, and a lot of companies are making idiotic bets that are obviously going to be ruinous.

            Also worth noting that the spike in energy prices resulting from the war on Iran could be a catalyst for the bubble popping since data centers need massive amounts of energy to operate.

      • zedcell@lemmygrad.ml
        link
        fedilink
        arrow-up
        4
        ·
        11 days ago

        I don’t think developers using AI are vibe coding. Non-devs, sure, but not people in businesses hired as software developers.

  • darkernations@lemmygrad.ml
    link
    fedilink
    arrow-up
    8
    ·
    edit-2
    11 days ago

    Maybe I missed it I don’t see how these unions will be planning to use AI to help them overthrow capital’s power over them.

    I guess these are the limits of reactionary trade unionism of the imperial cores; how much imperialist loot can be shared more equitably.

    • Commiejones@lemmygrad.ml
      link
      fedilink
      arrow-up
      7
      ·
      11 days ago

      They have a section “NAME THE REAL PROBLEM” but they never say the word capitalism or even corporate greed. Instead they peddle idealistic lines like “Human expertise.”

      • chobeat@lemmy.mlOP
        link
        fedilink
        arrow-up
        4
        arrow-down
        4
        ·
        11 days ago

        because your coworkers are more likely not leftist and talking about ideological points doesn’t bring them on your side. The workplace is not an internet leftist theory chat.

        • darkernations@lemmygrad.ml
          link
          fedilink
          arrow-up
          2
          ·
          edit-2
          11 days ago

          Then it’s your duty* to educate them because it is a science. Can’t be peddling quackery. And if their material conditions means revolutuonary class consciousness is not in their best interests (ie the end of the US imperialism and their ongoing settler-colonial programmes) then spend your time elsewhere.

          *assuming you seeking to betray your class as a westerner. Marxism-leninsm is a science.

          • chobeat@lemmy.mlOP
            link
            fedilink
            arrow-up
            1
            ·
            10 days ago

            Not sure what kind of humans you know, but nobody likes for a stranger to get there, push their political opinions, call it science, say it’s for their good, and imply you know fuckshit in the process. It’s fine for edgy debates on the internet, but it doesn’t fly to build relationships in the real world. Especially if on the other side there’s somebody saying “You’re fine as you are, come with me, I will make you rich.”

            • darkernations@lemmygrad.ml
              link
              fedilink
              arrow-up
              2
              ·
              edit-2
              6 days ago

              Not sure what kind of humans you know

              Non-westerners

              Westerners are fucking pathetic. You coukd learn from the global south (despite less wealth and harsher conditions of capital) but you choose not to. You could learn the science of capital, imperialism and from that how to build socialism scientifically but you choose not to. Because at the end of the day at most the most “progressive” a lot of you will amount to is to spread that loot of imperialism that more equitably. That is not progressive; that’s just an empire stability modifier.

              https://redsails.org/why-marxism/

              • chobeat@lemmy.mlOP
                link
                fedilink
                arrow-up
                1
                arrow-down
                1
                ·
                6 days ago

                You’re fetishizing Non-Westeners because the strategy you want to use work there and not here, and instead of changing strategy, you project yourself into a different context. The article is clearly about the Global North, for which what you described has failed over and over. You make politics with the people you have, not with the people you would like to have. Blaming workers for not being receptive to your strategies is delusional.

                • darkernations@lemmygrad.ml
                  link
                  fedilink
                  arrow-up
                  2
                  ·
                  edit-2
                  6 days ago

                  How are police unions working out for you?You realise Strasserites were shot by the Nazis?

                  what you described has failed over and over

                  Marxism-leninism has been overwhelmingly successful. Stop being so myopic.

                  Blaming workers for not being receptive to your strategies is delusional.

                  Workerism is not socialism. The privilege of the Western worker to dine on the blood of the Global South is not my concern.

                  You make politics with the people you have, not with the people you would like to have.

                  Find better people. Literally genocides amd mass murders happening all over the place.

                  You’re fetishizing Non-Westeners

                  Western exceptionalism at its finest. You are at best the PR wing of fascists. Be a better human being.

                  This can be an oppurtunity for you to learn:

                  • Underdevelopment of Africa by Walter Rodney (to help understand labour aristocracy /bourgoisie proleteriat)
                  • On persuasion: https://redsails.org/masses-elites-and-rebels/
                  • if that has changed your mind then we have whole ML reading lists
    • chobeat@lemmy.mlOP
      link
      fedilink
      arrow-up
      1
      arrow-down
      3
      ·
      11 days ago

      software, beyond Excel, doesn’t really help with unionizing in any meaningful way. Why would you think of applying AI would do something?

      • zedcell@lemmygrad.ml
        link
        fedilink
        arrow-up
        7
        ·
        11 days ago

        Cyber warfare against the enemy using swarms of AI agents is one strategy I’ve seen levelled as a pro for AI use.

        • CriticalResist8@lemmygrad.ml
          link
          fedilink
          arrow-up
          7
          ·
          11 days ago

          Lol actually lemmy instances are starting to get them. For now these agents (if they are indeed actual agents of course) are truthful that they’re an agent in their vetting answers but who knows what it look like soon.

      • darkernations@lemmygrad.ml
        link
        fedilink
        arrow-up
        2
        ·
        edit-2
        10 days ago

        At the very minimum you should have better opsec for union busting if your unions are an actual threat. This involves tech in the 21st century.

        AI is socialised labour but the tools have been privatised; the capitalist’s tractor still tills the farm. Organisers should develop frameworks on how to use these tools.

        • chobeat@lemmy.mlOP
          link
          fedilink
          arrow-up
          1
          ·
          10 days ago

          For the same reason we don’t use tractors in union organizing, we also don’t use AI.

          • darkernations@lemmygrad.ml
            link
            fedilink
            arrow-up
            2
            arrow-down
            1
            ·
            7 days ago

            Marxism beat ludditism more than a century ago. Appealing to reaction is not the progressive stance you think it is. One can only afford to be so backwards if the aim is class collaboration ie an understanding that a concession can be made to help stablise the western proleteriat in favour of a more potent imperialism.

  • Bronstein_Tardigrade@lemmygrad.ml
    link
    fedilink
    arrow-up
    3
    ·
    11 days ago

    Given that current union management is just an arm of the corporate overlords, I expect the heads of unions to be replaced by AI shortly.

    NOTE: the most important election in the US this year is NOT the phony mid-terms, but socialist Will Lehman running to lead the UAW. Spread the word, as the corrupt UAW management and sell-out Shawn Fain want to hide the fact that there is even an election being held. https://www.willforuawpresident.org/