https://www.geeksaresexy.net/2023/09/04/why-the-growth-of-ai-in-making-art-wont-eliminate-artists/
Will AI replace artists? On one hand, the answer to whether AI will replace artists is no.
Generative AI is a powerful tool that can expand the possibilities of art making and will still require the guiding hand of a human artist. As with any new technology, some creative processes will become both easier and less time-consuming with AI.
I’m aware of where I’m posting, but the joy of making art is as much in the process as the result. Give it a short while and we’ll have ai that makes prompt generation even easier. Then even less of the individual will be in the work. At what point do we say, “you’re barely doing anything here”?
Let me ask you this: Would you consider photography art? And if so, is it fair to say that photography is a cheap form of art because the process is often less intense than putting a brush to canvas?
See, this is how I know you don’t know anything about the creative process. You think photography is just clicking the shutter. Not composition, lighting, shutter speed, etc.
Plus development is a science.
It’s not the same as asking a computer to copy someone else’s work.
No need to be hostile. I know plenty about creative process, including photographic process. What I’m asking you is, if art is directly tied to the process of creating it, in the same way that photography has a different process than painting, would ai art be considered art to you if you learned that there actually is a process of refining the product?
That is to say, you pointed out that there is a difference between clicking the shutter and taking light, composition, shutter speed etc, into account. One of those produces an ameteur product, and one produces a photo worthy of a gallery. And the difference is the artist.
So if you were to learn that not all ai art is even close to art, that in fact the vast majority of it is akin to the amateur clicking the shutter, would you consider good ai art any more valid given a process you may not be aware of?
No, because a computer cannot create. It cannot make design decisions.
Yes, it is a tool. A camera is a tool, and yet there is a process in which it is used to create art. A paint brush is a tool, but it’s not the thing that decides what goes on the canvas. A tool does not create without an artist.
What you’re arguing, that a computer does not create, isn’t even the point. Ultimately ai generation, like any tool, is only as good as the vision of the person piloting it.
It’s pretty clear to me that you don’t want to hear what I’m saying anyway. You want to be angry about a tool.
Good luck with that 👍
I would say photography is absolutely art, but the art of what? It’s the art of physically capturing and at times manipulating light. It involves elements of composition, contrast, timing, and a deep understanding of how these elements comes together to create or capture a desired effect.
I understand the parallel you’re making, but I think AI art is substantially less skilled medium than drawing, painting or photography. And over time, the amount of skill required will be even less still. What unique skills does it require? I’m not saying it requires no skill at all, but it’s strange to speak of a visual medium where you don’t really need to understand artistic fundamentals because it’s largely taken care of by the algorithm.
I certainly believe that a trained artist could achieve better results more quickly with AI, but the difference between an expert and a total amateur will narrow over time. Whatever coaxing is required now will get simpler. You could ever make a similar argument with photography. Someone with a modern phone can take much better pictures with much lower effort than in the past. What if you were to combine that with an ai which suggested ideal compositions based on a dataset of high quality photography? That would be cool tech, it may even be an interesting training tool, but would it mean that the people using it were skilled photographers in their own right? It’s similar with AI generated art.
So you would say that the tool devalues the artist?
I know it’s not so simple as that, but in essence the argument that you’re making is that as technology improves the baseline technique available to beginners, the gap narrows and devalues the work of an experienced artist.
In my opinion, that’s backwards. Art is subjective by its very nature. People using ai to generate art without a background or technical expertise in that art aren’t artists in the same sense that a person building a house in the Sims isn’t an architect. Does the Sims allowing a player to design a house devalue the job of an architect, whose purpose is to create something actually valuable? I think a architect would laugh at that notion.
In the same way, an artist creates value in a way that a non-artist cannot, even with ai generation being a factor.
I wouldn’t make that specific statement, no, but I do have sentiments that might be considered adjacent to that.
Suppose you have a watercolor artist who has spent years understanding their medium and mastering it. Then you have an ai artist who gives a prompt (simplifying here) of “watercolor painting of X in the style of X” and gets a result that looks even 80% of the way there. In the eyes of average people, the ease of achieving that result does affect in some fashion how they look at the work of the trained artist and how they value it.
We have examples from one of the key Dungeons & Dragons artists for instance where he has said in no uncertain terms that ai art has absolutely diluted the value of his work. People create ai works in his style and people can no longer tell what’s his and what’s ai. Then poor artistry in the ai works gets attributed to him.
The speed with which people can create ai work means that we will increasingly be buried under a glut of mere content and most creative work will be trivialized and devalued except for those who themselves have the skill to recognize what goes into skilled art.
I understand what you’re getting at, but that doesn’t mean that that artist’s work is any less valuable. In your example, the artist in question is still being paid to create genuine art. I suppose there is an argument there that with ai available to create similar art, he may otherwise lose commissions to lesser quality substitutes. But on the other side of the coin, it’s possible that many people emulating his work would accept a much lower quality substitute for convenience and wouldn’t have paid him to create anyway, particularly given that d&d art is often used for conception and token purposes and not quality.
On the topic of ai art affecting that artist’s reputation, I can see how it might become a problem that certain ai pieces are incorrectly attributed to an artist. That problem seems largely unavoidable though to people unwilling to differentiate or inquire as to who the artist is in the first place. That might also be something subjected to copyright law in the near future, like an ai water mark being required on any ai generator.
I still don’t think that creative work is even going to come close to being trivialized. Ai art is free to the public, and yet I still know several artists who live off of commissions, one of which is my wife, because ai can’t yet produce what people actually want to see with the refined quality that they can get from a real, human artist.
I think people will still find joy in it, just of a different sort. As an analogy, used to be that you had to program a computer with assembly language, and it was rather painstaking to do. There were some real wizards doing that sort of work, like in the Story of Mel. Nowadays though, we’ve got high-level languages with compilers that do all the grunt work of actually writing assembly for you. Some people still worry about assembly specifically, but the vast majority of programmers don’t. The joy is no longer in writing the absolute fastest bare metal assembly you can for most people, it’s in using the right algorithms to solve a problem. You can write a few lines in Python that would’ve taken someone weeks to write in assembly.
Something relevant to note in the above is that smarter compilers eliminating the need for people to write assembly directly didn’t mean the end of programming as a profession. There’s been an explosion in programming jobs, exactly because each person can do so much more than they ever could before, opening up new possibilities that weren’t there with everyone writing assembly.
Likewise, I suspect the scale will change here. I admit I’m not an artist, but wouldn’t it be cool to see your artistic vision across an entire game? You could create an entire virtual living breathing city on your own without having an army of artists working on the grunt work like the exact concrete texture. If you decide to tweak the feel of the art, you don’t have to spend weeks redoing all the grunt work. Alternatively, if you draw a landscape that you get just right, imagine being able to experience it by having the AI generate a virtual world for you to walk through based on it.