Raw LLM Responses

I'm not a tech bro and while I understand the argument about AI being "theft" there is one problem with that argument that I yet need to be convinced about, to me it shows actual misunderstanding of how involved using AI is. When humans learn their skill, they also copy what others had done. They get inspiration and experience by analysing art of others, and in a way, they do exactly the same thing as the AI, the only difference is that it's done by a human brain and manual control over the pencil, instead of being generated digitally. But really, it's still generated based off pre-existing things, but by human. This is also why plagiarism is a huge topic because the border between when a human commits plagiarism and when they did something different enough to be considered original is very subjective and even when some countries try to outline rules in the legal system, these kinds of court cases are always very difficult and take years. So really, whatever humans do is really no different for as long as AI is not plagiarising. And you can't say that AI is always plagiarism if you allow humans to copy someone's artstyles and otherwise take inspiration, especially that it's relatively rare with AI that you ask it to create a copy of an existing work, but usually you will pick different loras and so on to combine them into something new that hadn't existed before in this particular combination, there was still the element of human curating the training data, also the creative element when writing prompts, and so on - it's just as much creative as a person drawing a picture, really, but people are prejudiced because the processing abd generation was delegated to the machine - the part of the process that actually has nothing to do whether a given work was made in a creative process or not, whether it's plagiarised or different enough to be considered original. I yet need to hear someone refute that without gaslighting and dismissing the creative aspect of curating datasets (basically the same as artists looking at works of others, with datasets being created in their brains), combining them together to be able to make something new that hadn't existed (literally artists are being called innovative when they blend pre-existing styles, but when it's AI it's somehow still theft), and dismissing the aspect of introducing a new creative idea (prompt) being given to the machine that just does the manual work for you (the drawing). Why is it okay to do these literally the same thing manually, but when you use tools to help you with it, it's somehow theft. It's not like humans using other artists' work as inspiration, studying them to hone their own craft, ask for a permission to do so, so why is it okay to do it if AI is bad? To me, using AI is a next step to what invention of digital art was where to create a gradient of colours software could sort that out for you instead of you needing to do all the in-between colours manually. It IS a tool, just a more elaborate one. And what is going to be happening is that generative AI will just keep evolving into a more assistive role, allowing people to have more and more control over the process, it will be less and less of a lottery of the result you get. It will become just a different type of skill that will exist in parallel to regular art, and images made with today's generative AI with it's randomness, misunderstandings of the prompt and artifacts, will be an equivalent to drawing stick figures, something rookie stuff. I don't really see a reason to hate that, how is that any less creative, especially once people will have control over where a given stroke is being placed when they can re-edit a given piece with further prompts to make the AI give output that was actually what was intended behind the prompt. We're just at early stages of that, so the tools aren't as advanced yet, but they still can be used in a very much creative way that is controlled and deliberate to a massive extent, with human internationality being behind it. It's just that the skill is to know how to make the machine do what you actually want it to so, as opposed to training your brain and hand to have that skill in case of drawing and painting, and in case of digital art that is already somewhere in between, especially when the software will be correcting the shape of your strokes and then copying and pasting them, flipping around, resizing, stretching, and filling in colours and gradients and so on, as opposed to doing everything manually... Still relying on the machine to make a portion of the work. And AI is serving the same function, just going a few steps further with it and it's a matter of time for the controllability of it to improve. Preparing datasets properly and training AI and then also training yourself how to prompt may not take years, sure, but it can still easily be months before you start getting what you actually wanted, it takes an actual learning curve how to train and use AI... So what gives.