To be honest, I’m pretty sure I already know the reasons but I’d like to see other people’s takes on it.

  • definitemaybe@lemmy.ca
    link
    fedilink
    arrow-up
    0
    ·
    4 hours ago

    I skimmed all the comments, and saw two that touched on the real answer, but didn’t fully answer it:

    Tech companies are “growth companies”. This means that they’re valuation is based on expected future exponential returns, not current profits.

    But they’re monopolies that have already captured almost the entire market worldwide. There’s no growth left in their traditional businesses.

    This is why Meta is now “Meta”—Zuck needed to convince investors that VR was the “next big thing” to justify his company’s valuation. In reality, Facebook is more like a utility company. They sell (a lot of) ads. But that’s a relatively stable business, and they’ve saturated the market.

    So, now it’s AI. It needs to be big, or else the companies aren’t worth their share price. And executive stock bonus options aren’t worth the paper they’re printed on.

    That’s my take.

  • MyBrainHurts@piefed.ca
    link
    fedilink
    English
    arrow-up
    0
    ·
    8 hours ago

    Depends on the context. The big companies betting heavily on it, need it to work. Hence the push by Microsoft, facebook, google etc. And those make up a huge part of the normie tech experience.

    For competitors, the concern is that a lot of people are using it in different areas. I hate the ai but sometimes it’s just more efficient and easier. So, a browser like duckduckgo or Firefox has a dilemma, the folks who enjoy ai probably want it incorporated into their experience. Those who don’t, can usually turn it off.

    Trivial example, I wanted to watch the Avery/Patrick Stewart episodes of American Dad. There’s some weirdness with the episode ordering so when I was checking google links, it would take a few tries per episode. By the third time, I glanced at the ai summary I’d skipped and sure enough, it listed both possible episode #s.

    And then of course, for companies, the developers are a pretty heavy chunk of the budget. If proper ai use can get them producing faster, that’s a no brainer. (Whether the increased production bests the bugs etc is a different story.)

  • anticurrent@sh.itjust.works
    link
    fedilink
    arrow-up
    0
    ·
    9 hours ago

    Because they are in a race against time to reach AGI “Artificial General Intelligence” before a market crash and investments drying up.

    The end goal is the be in control of the biggest and strongest weapon on earth, become monopolies and garner wealth like never seen before

  • hightrix@lemmy.world
    link
    fedilink
    arrow-up
    0
    ·
    10 hours ago

    What’s the easiest way to get personal or private information about a person? Ask them to give it to you.

    First Facebook asked people to post all their private details in public. Now AI asks people to give complete access to your information.

    We thought the social media driven ad market was bad. Just wait…

  • 🇨🇦GreenBeard🇨🇦@lemmy.ca
    link
    fedilink
    English
    arrow-up
    0
    ·
    10 hours ago

    Monopolies/Oligopolies are the most likely answer. It’s like most things these days, they function on an addiction model and then make you pay a subscription once you’re dependent on it. Opioids, streaming platforms, social media, Amazon, vaping pens, it’s all about the “Recurring Revenue Streams” AKA profiteering off of cognitive biases, network effects, and chemical addictions so they can capitalize on infrastructure monopolies.

    Frontier AI models require massive amounts of computing power to train which puts them out of reach of average people. Oh sure, you can run a local open-weight release model on a decent gaming computer, but you can’t really roll your own and the best stuff is locked behind a paywall. It only works if people want it though, so they’re spamming it in everything and hoping people will find something to use it for that in a few years they can’t live without, so they can monetize it.

    They’re just throwing it at the wall to see what sticks. That’s the marginally less insidious part.

    There’s also the fact that the more you use it, the weaker your own analytic skills get. It’s like how distance running used to be the one physical trait humans were honestly pretty good at once upon a time (well, that and throwing things real good). Now people who run marathons are weird and most of us couldn’t get to the store down the street without a car, an uber, a train, or at least a bike or a scooter. Bonus points, if you can’t think without the machine, you can’t think of a way to get away from the machine.

  • ooli3@sopuli.xyz
    link
    fedilink
    arrow-up
    0
    ·
    11 hours ago

    that’s a good question, too bad nobody, in top reply, didnt make a tentative to give an honest answer

    My guess in order of what I believe:

    • Free Data. As any new tech since google: Using free “work” from user (any interaction with LLM is improving its general output, like any search request on google was bettering its result) , to enhance the model. They need real human data and text more than ever now that LLM generated content is polluting all website, and the best way to get them is offering “free” llm.
    • Stolen data: Getting people data, to better target ad
    • Try to get people hocked. so they would end up paying for it, else they make no money
    • Get people confortable with it, so they would push the company they work in, to use it. Window approach. They would offer LLM in school in university real soon, if not already
  • some_guy@lemmy.sdf.org
    link
    fedilink
    English
    arrow-up
    0
    ·
    14 hours ago

    GPT-3 was genuinely surprisingly cool and blew a lot of peoples’ hair back. That led to FOMO from all the other tech companies. They all spent a ton of money on their own R&D and the media went with it without questioning the value, so it went unchecked. Now they’re all super over-leveraged and they need to dig themselves out. It’s gonna be a bloodbath when the bottom falls out.

    • ArmchairAce1944@lemmy.ca
      link
      fedilink
      arrow-up
      0
      ·
      12 hours ago

      To be fair most of my questions with that early model were silly questions like ‘why won’t Bill Gates buy me lunch?’ Or some such nonsense.

  • Swordgeek@lemmy.ca
    link
    fedilink
    arrow-up
    0
    ·
    14 hours ago

    Because fascist assholes want to use it to get richer and more powerful, and don’t give a shit about destroying the planet or humanity (in the long term), nor about making live objectively worse for everyone else.

  • mrmaplebar@fedia.io
    link
    fedilink
    arrow-up
    0
    ·
    14 hours ago

    Sunk cost fallacy. People have put so much money in it (the biggest financial bubble ever, by far) that they can’t afford for it to do anything less than replace most of the world economic. So now they are throwing shit at the wall and seeing what sticks.

  • Wytch@lemmy.zip
    link
    fedilink
    English
    arrow-up
    0
    ·
    14 hours ago

    Big Tech CEOs all want in on the Next Big Thing. No matter what is, or how valuable it turns out to really be. Regardless of the risks or the impact on anything and anyone. They are no longer just focused on making money, either. They’re well past that point. Now, the focus of the gilded class is to create a Legacy. To be the next Steve Jobs with the iPhone. To shape the Future.

    To that end, they’ve locked onto sci-fi and fantasy notions to guide them towards whatever strikes them as Cool Futuristic Shit. They’re not really visionaries, they’re poseurs. Idea thieves. AI has a firm place in many fictional concepts of what The Future will look like and how it will shape our collective fates. The dream therefore, to create an artificial mind, has been co-opted by an industry capable of building a sophisticated autocomplete engine.

    The final step in this process is to monetize this technology. Its use and maintenance is exceptionally costly however and remains of dubious or limited value, with little actual demand. The hype train is put into overdrive, selling this technology to any partners they can convince, consumers be damned. The goal of profitability is always looming. We are in the midst of this phase, with delusional billionaires vying to be the Guy Who Changed Everything, so they can justify their shitty desperation and lonely isolation from reality.

    They will continue to shove this crap into everything possible so the bubble keeps expanding, hoping to unlock some secret of the universe or at least, the threshold of profitability.

  • aramis87@fedia.io
    link
    fedilink
    arrow-up
    0
    ·
    14 hours ago

    The purpose of AI is to allow the wealthy to access the benefits of the talented, without letting the talented access the benefits of wealth.

  • HubertManne@piefed.social
    link
    fedilink
    English
    arrow-up
    0
    ·
    15 hours ago

    the commercials I see really crack me up. Either they are for such paltry things its like why bother using ai or its such complex things that is a sort of fairy tale of it instancely doing things that it just could not as portrayed.

  • FaceDeer@fedia.io
    link
    fedilink
    arrow-up
    0
    ·
    15 hours ago

    When the Internet first became “big” during the dot-com boom, why did so many companies rush to put up websites and do “internet everything?” Because it was obvious that there was huge opportunity to be had with this new technology but it was unclear which of those opportunities would pan out. So they threw everything at the wall to see what would stick.

    Same with AI. It’s obviously a huge opportunity. But we haven’t “done” AI like this before so it’s unclear which uses will be the big winners. So everyone’s trying it out for everything.

    • shalafi@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      ·
      11 hours ago

      Boom! AI will work out in some form or another and these companies know this is a monstrous bubble, but they’re banking on being one of the few who crawl out the blast crater.

    • Awesome Lowlander@cozyver.se
      link
      fedilink
      English
      arrow-up
      0
      ·
      14 hours ago

      Yeah so many on lemmy refuse to accept that there are some use cases for AI, and just blindly go AI bad about everything.

      • classic@fedia.io
        link
        fedilink
        arrow-up
        0
        ·
        13 hours ago

        My sense, for some of those people, is that the valid use cases of ai are dwarfed by the environmental and societal costs, not just ai is bad

        • Awesome Lowlander@cozyver.se
          link
          fedilink
          English
          arrow-up
          0
          ·
          13 hours ago

          I can understand that. But most are just complaining about ‘slop’, which is something else. Most of them at this point are just jumping on the bandwagon without understanding any nuances.

          • BurgerBaron@quokk.au
            link
            fedilink
            English
            arrow-up
            0
            ·
            12 hours ago

            without understanding any nuances

            Common issue with a lot of people. Sometimes it’s willful black&white thinking and there’s no conversation to be had with them. I’m deeply pessimistic in this very post elsewhere and yet I still think there can be positive use cases. Crazy, right? My dark grey will still offend someone who only wants to see black.

  • Hegar@fedia.io
    link
    fedilink
    arrow-up
    0
    ·
    15 hours ago

    Because ai companies think that more data and more users will train their product better to the point that they can win/rule the world/get all the money/etc.

    That’s the motivation for jamming it in everywhere regardless of how useful or safe that actually is.