Raw LLM Responses

Here's the thing I think everyone's doing without realizing. You guys talk about how the AI was trained on data that was copyrighted (which, while I would argue is a necessary evil considering how incredibly important AI such as these are to our future, is a fair point, I'll admit) but what I see you guys doing, whether intentionally or unintentionally, is going back and forth between that point, and what essentially boils down to "its taking our jobs." And that second point is just, so very problematic. Yes, it is. And no matter how saddening it might seem to you guys, even if the AI starts from scratch again using data-sets that are squeaky clean legally, it will become just as good as it is currently, if not far better, within just a couple of years, at the maximum. *And if not now, then by that point, your jobs will be gone*. And yes, its not fair. But it wasn't fair to telephone switchboard operators, when automatic systems completely eradicated that industry.  Would you have preferred that they campaigned to prevent these systems from becoming commonplace, talking about their jobs being taken, as you are now? *_Would you have supported them, and accepted a much less convenient, much less advanced society as a result?_* It is easy to not see the big picture when you're the one suffering, and I'll admit, easier for me to lecture you about it, being a programmer myself. But I implore you, think about that scenario, and what you would have done. Secondly, you can cry unfairness, and talk about art being creativity and human touch and all of that, but the results of the ML models speak for themselves, and people like them. *People, even trained artists sometimes, cannot reliably differentiate between AI generated images, and human ones*. And that level of visual fidelity is what you get at the infancy of this technology. Consider then, that if, to an end-user, there's not much of a difference between what your years of work and creativity amount to on a canvas, and what they can get out of an AI model in 10 seconds, that then they're going to choose that 10 times out of 10. *That is not bad, that is just logical*. AI art is the future. Because it looks good. Because it takes very little time to make it. I've heard artists talk about these exact points as if they're _bad_. "AI art will never be equal to ours because those machines produce a hundred different pieces soullessly while create each of mine slowly, and with care". I'm paraphrasing, yes, but this is hilarious to me. Its like saying cars will never be equal to horse drawn carriages, because cars can go so much faster, are so much more efficient, and so much more accessible. Really? You're making our point for us. Also, as for creativity and soul in artwork, people looking to use artwork commercially do not, and *have not ever* cared about that (nor do 99% of the people who consume your artwork). Companies especially, which hire the largest amount of artists only care for art that looks good, and that fits a theme, and that is affordable and efficiently gotten. AI art fits that category, and has, in its infancy, already outclassed and outstripped you in every possible metric. That is not a problem, that is literally why it was made. Its a selling point. That is the future, and is inevitable regardless of whatever you may feel or do now. If you truly do care about soulful, purposeful art, though, then by all means, keep creating! AI art isn't slapping the brush out of your hand; no one is preventing you from putting color on the canvas. Artists talk about art as a medium of self-expression; express yourselves then. You won't ever be able to compete with AI in what it offers as a service, but that isn't something to be angry about. Hell, working in an office creating posters isn't what drove you to art in the first place, right? So what's changed? You're not angry you can't transport people on your back faster than a car, right? Same principle. Also a tiny point. I hear people talking about AI models spitting out exact replicas of artworks a lot. Firstly, that is *highly* unlikely to ever occur. Impossible, really. Even if you specifically ask for exact replicas, it is actually a problem that the AI models have tiny differences in even providing consistent results to an identical prompt across different requests. So an exact replica of one of its images across a dataset so large is simply not possible currently. People are *trying* to do it, and it still can't be done yet. Lastly, I find it incredibly haughty that some people here are seriously talking about getting rid of AI models altogether. You may be of the art world, and perhaps not know this, but the baseline code for the more popular models, Stable Diffusion, for example, is free, open-source (meaning anyone can look at and copy or distribute the code) and publicly available, with probably a million if not a billion separate full copies of it floating around on the internet, and on offline storage mediums in people's computers.  What this means that, if tomorrow, every single major government on the planet decides to totally ban all forms of AI generated imagery, completely, even that will not be enough to proverbially put the cat back in the bag. It is here. It is permanent. Any reasonably motivated individual can create their own custom model from available sources within a month at maximum. Regardless of how you feel of it, that's what it is. And it is never going to change. Logically, why should companies or even individuals pay you now to make something unique, when they can simply request an AI to make whatever their heart desires, in a thousand different variations? And I do mean unique, beautiful artwork, never before seen, generated by an AI that is happy to change it infinitely many times, to exactly what you want. Involving an artist in it would be like rubbing stones together to produce fire, instead of using a lighter. It would be foolish and utterly useless in every way. There's just no point to it.