The fediverse used to feel pretty anti-ai, but over the past month or two I’ve noticed a LOT of generated memes and images, and they tend to have positive votes.

Has there been a sudden culture shift here? Or is there a substantial percentage of people just unable to tell the difference anymore?

  • Cris@lemmy.world
    link
    fedilink
    English
    arrow-up
    2
    ·
    edit-2
    1 day ago

    Is that generative ai or machine learning? I really don’t have the same issue with machine learning

    Its not just llm’s, I find calling ai generated images “art” frankly offensive and demonstrates a fundamental misunderstanding of what art is and why it’s important to us as humans. But my issue is really just generative ai

    I don’t really think people have an issue with machine learning, it’s useful for all kinds of stuff, and doesn’t really come with the same ethical problems as best I’m aware, so I have no reason to complain about it.

    • Valmond@lemmy.world
      link
      fedilink
      arrow-up
      1
      arrow-down
      1
      ·
      1 day ago

      AI is machine learning. It was just called “machine learning” before every C-suit had to try to sell it.

      I’m with you when it comes to shitty images, but there is interesting approaches too IMO, especially when they will get better, if they do. As for art, it’s just another tool in the toolbox. Painters treated photography similarly, and do you remember (if you’re old enough) when digital cameras became affordable and everyone and their grandma became “a photographer” and flooded the planet with soulless photos? I do 😅

      Art is art, no machine will change that, but maybe it will help people get into the arts, with the cost of a lot of slop ofc. Which is cool IMO.

      • Aceticon@lemmy.dbzer0.com
        link
        fedilink
        English
        arrow-up
        1
        ·
        edit-2
        10 hours ago

        I work in Game Making.

        There’s a ton of stuff in it which has been called “AI” for literally decades and almost none of it is Machine Learnrning: for example the A* pathing algorithm for characters in a game is called “AI”, as are Steering Behaviours that can be used in things like simulating bird flocks, and both are entirelly algorithmic, not ML.

        In my own experience ML is seldom useful in games, mainly because algorithms are lighter and generally work more reliably.

        You’re confusing use of “AI” in the Marketing of the present day tech bros trying to make money pumping up a Tech Bubble on top of certain very specific forms of Machine Learning, with the actual general meaning of the acronym.

        • Valmond@lemmy.world
          link
          fedilink
          arrow-up
          1
          ·
          10 hours ago

          I have 10 years of game dev under my belt, and I wonder where your idea that I think A* is AI come from.

          Do you know how machine learning works? Please tell me about this new AI that isn’t based on it!

          Don’t get me wrong, I think you’re not wrong, Dijkstra isn’t AI, techbros are trying to shovel “AI” into everything for the buzz, but you’re explaining yourself a bit haphasardly IMO.

          • Aceticon@lemmy.dbzer0.com
            link
            fedilink
            English
            arrow-up
            1
            ·
            edit-2
            9 hours ago

            Because shit like that has been called “Pathing AI” for ages.

            (For example)

            Also I’m very familiar with Machine Learning having actually learned it 3 decades ago when it was mainly just Neural Networks (there were other techniques schu as Genetic Algorithms, but ultimately NNs became dominant and is most of what we today call Machine Learning) and its most advanced commercial use was to read postal codes in mail envelopes for automated mail sorting.

            The acronym AI has been thrown around for decades, even before Neural Networks were invented and well before Machine Learning was even called “Machine Learning”.

            • Valmond@lemmy.world
              link
              fedilink
              arrow-up
              1
              ·
              9 hours ago

              Yeah the MNIST dataset, fun times.

              I don’t even know if we’re on the same page or if there is something we dabate, everything you say is totally valid IMO.

      • Cris@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        ·
        edit-2
        1 day ago

        Generative ai and machine learning are pretty broadly considered different if adjacent technologies if I’m not mistaken

        https://www.forbes.com/sites/bernardmarr/2024/06/25/the-vital-difference-between-machine-learning-and-generative-ai/ probably not the best source, I just grabbed what came up

        They don’t really do the same thing, and have different types of outputs compared to one another, even if both use a neural network of weights or whatever

        As a person who has spent a huge amount of my life making art I think the idea that it will get more people into art is naive, and I think being devoid of understanding artistic principles it makes poor reference compared to anything else, which is part of why artists communities loathe generative AI. I follow tons of artists online and they all periodically have to stop and vent their frustration.

        A youtuber artist did a whole video explaining how finding reference on the internet is now borderline impossible due to ai content, and after problem solving explained you can avoid that problem by only looking at images older than when gen ai became widespread. It reached a pretty big audience and was extremely well recieved by artists, broadly, hate gen ai and want nothing to do with it 😅

        Also just wanna clarify- I’m not downvoting you. I try to downvote when a comment is bad behaviour or doesn’t add anything, not just when I don’t agree with someone

        • Valmond@lemmy.world
          link
          fedilink
          arrow-up
          1
          ·
          1 day ago

          Interesting, my art friends, around 50% just love using generative AI for images to use, the other just don’t care. No one does digital art though, or not as their main style for what I know (acrylics, oil paint, gouache, aquarelle etc ).

          I don’t like AI gen images because there is always tons of crap in colors and lighting, I guess for composition it could be useful but I don’t use images for that personally, so I like go on pinterest.co.kr with an ad blocker and use my imagination.

          Yeah there is someone on a downvote spree lol, no problems & thanks for your post!