>Microsoft invested $10 billion on this

>Microsoft invested $10 billion on this

Beware Cat Shirt $21.68

Rise, Grind, Banana Find Shirt $21.68

Beware Cat Shirt $21.68

  1. 1 month ago
    Anonymous

    An executive made a big speech to investors about AI as some emerging thing, earned a fat bonus, then fricked off to another company before it came out that the money was all wasted. 99% of corporate projects are this same shit repeated.

  2. 1 month ago
    Anonymous

    Look, there's a few kinks, but it's still revolutionary and worth the investment. Stop b***hing about experimental developments which will inevitably improve over the long term. You're not being witty or clever about a few tiny errors. You're just a jackass. Grow the frick up, butthole.

    • 1 month ago
      Anonymous

      >total failure
      >* a few kinks*
      Microsoft Executive detected

      • 1 month ago
        Anonymous

        how is this a total failure?

        • 1 month ago
          Anonymous

          an AI that can't count to four.
          go on kiddo, push that braincell.

        • 1 month ago
          Anonymous

          It can't do basic math.

          • 1 month ago
            Anonymous

            if it hasn't seen the problem before it has no idea how to handle it. garbage in -> predicted garbage out with % accuracy.

          • 1 month ago
            Anonymous

            It hasn't "seen" any "problems" though.

          • 1 month ago
            Anonymous

            >if it hasn't seen the problem before it has no idea how to handle it
            Then it hasn't learnt how to adapt and therefore is not intelligent. Why would I need something that only solves problems I have already faced?

    • 1 month ago
      Anonymous

      >, but it's still revolutionary
      only moronic monkeys that have no idea how this technology works call it "revolutionary". ai memeshit algorithms for image generation isn't black magic only shitcoin grifters pretend to know about.
      > rest of your cope
      kys thanks.

    • 1 month ago
      Anonymous

      The eventual future of AI is a tool that speeds up productivity for people that already know what the frick they're doing. The current craze is a bubble that's fueled by sunk cost fallacies and the belief that you can make the singularity happen if you just throw enough GPUs at the problem, it's not going to last

      The problem is, AI, and especially Neural Networks, are a black box
      No, it isn't easy to "improve" and "remove a few kinks"
      No, it isn't east to get it to integrate the interface with humans to enable it to be a productivity tool
      It will get stuck in local optima like this, and then fail to improve

      • 1 month ago
        Anonymous

        While the neural weights themselves are a black box, you can still train it to act in a certain way using human reinforcement learning. By feeding it input, and teaching it which output is bad and which is good. This is how OpenAI trained ChatGPT to become woke.

        • 1 month ago
          Anonymous

          This doesn't improve its capabilities, it just changes its outputs in some specific circumstances, and likely makes it worse overall because NNs have a tendency to forget what they previously learned when they're tampered with

  3. 1 month ago
    Anonymous

    Just edit the damn image with Photoshop you moron

  4. 1 month ago
    Anonymous

    Midjourney gave me 50% correct on the first try.

    • 1 month ago
      Anonymous

      midjourney is 200% better than dall-e

    • 1 month ago
      Anonymous

      Odd that it won't just give you two red two white but insists on doing a gradient.

  5. 1 month ago
    Anonymous

    i believe this has to do with their ambiguity filter. they don't want you having full control of the characteristics so they can inject diversity

    • 1 month ago
      Anonymous

      just be glad there isn't a black woman's hand pouring wine into one of the glasses

  6. 1 month ago
    Anonymous

    It's interesting AI has turned out to be good at creative interpretations but shit at anything logical, the exact opposite of science fiction ideas.

    • 1 month ago
      Anonymous

      your takeaway should be that glorified google image and reddit searchers arent AI

    • 1 month ago
      Anonymous

      You only think that because you're not very creative but are very logical, so you can see clearly when it bullshits and fricks up its logic.

      Ask any artist about the shit that generative "AI" sharts out, and they'll tell you it's fricking garbage and be able to explain exactly why.

      Generative "AI" is shit at everything except convincing people that don't know better otherwise.

      • 1 month ago
        Anonymous

        no I am not buying your furry porn
        you will be replaced

  7. 1 month ago
    Anonymous

    I had a talk with my brother today that told that in Belgium right now they're making diplomas specialized on AI.
    Is IQfy the only part of the world that don't believe in AI's future? Because I was aggreing with all of (you)s but now I'm not so sure anymore.

    • 1 month ago
      Anonymous

      Most countries are still messing up on their digitization and they think that being early on AI will save them since they falsely believe that AI will cause a massive shift in the job market. As you can clearly see it's not there yet and it's doubtful it will ever reach the point of being used outside of a few niche use cases that can be fulfilled by already available tools.

    • 1 month ago
      Anonymous

      The eventual future of AI is a tool that speeds up productivity for people that already know what the frick they're doing. The current craze is a bubble that's fueled by sunk cost fallacies and the belief that you can make the singularity happen if you just throw enough GPUs at the problem, it's not going to last

    • 1 month ago
      Anonymous

      >Because I was aggreing with all of (you)s but now I'm not so sure anymore.
      It can take 10 years for a fad to die.
      You can even make a small fortune chasing fads, making a ton of money in the bubble.
      Just once the fad dies, people will either have to retrain skills, remain a grifter for life, or retire on a small fortune made during the bubble.
      Just like Bitcoin millionaires.

    • 1 month ago
      Anonymous

      What is there to specialize in? Either you are a comp sci phd, a dev working the data or devops pipeline, or an end user.

  8. 1 month ago
    Anonymous

    >Microsoft invested a small portion of $10 billion dollars in this and used it to launder the rest

    How could they possibly spend $10 billion? Think about that amount. They could spend $100 million on real estate, equipment, staff, fricking ping pong tables, then spend that 10x over, then spend THAT 10x more.

  9. 1 month ago
    Anonymous

    >using GPT-3.5 as an image model
    holy frick IQfy is moronic I hate this board

  10. 1 month ago
    Anonymous

    I feel like it's not actually making mistakes, it's trolling us from our 3D mindset and this AI already went AGI and is just casually posting its non-euclidian generated images so it's true but we cannot comprehend why with our 3D minds.

  11. 1 month ago
    Anonymous

    one of those cups has applejuice

  12. 1 month ago
    Anonymous

    If the prompt fails the first time, start a new page. Telling it made a mistake and correct it causes it to lose confidence or something and make way more mistakes.

  13. 1 month ago
    Anonymous

    sounds like skill issue

    • 1 month ago
      Anonymous

      >takes 7 hours of proooompting
      just stop
      GPT is a text model not an image model
      you shouldnt be using GPT for this anyway

  14. 1 month ago
    Anonymous

    Chatgpt reminds me of the offshore Indians guy I work with.r0yjh

  15. 1 month ago
    Anonymous

    you didn't tip your AI enough

  16. 1 month ago
    Anonymous

    it got one out of four right for me on the first try

Comments are closed.