Renderosity Forums / AI Generated Art
Welcome to the AI Generated Art Forum
Forum Moderators: wheatpenny, RedPhantom, Deenamic
AI Generated Art F.A.Q (Last Updated: 2024 Aug 27 11:07 am)
Dream, by Wombo https://dream.ai/create
Subject: The good and bad things about AI generated images?
Before i even bother to take the time to reply to this post, and its many points, have you actually used AI... Seeing as it is that easy to use and you can turn out images of very high quality in a few minutes, i would like to see who i'm replying to and their understanding of how it work, please show your AI creations here and your understanding of the medium...
Before i even bother to take the time to reply to this post, and its many points, have you actually used AI... Seeing as it is that easy to use and you can turn out images of very high quality in a few minutes, i would like to see who i'm replying to and their understanding of how it work, please show your AI creations here and your understanding of the medium...
I'm not sure how that will be relevant to what I wrote, but here are some of the stuff I made, using different styles.
I'm sorry, but I don't find your examples to demonstrate high-quality images. Given you have chosen to make close-up images and you seem to have aimed for realism, you should have made sure that the eyebrows and hair look realistic as well, as they stand out as being very fake, also the lips seem a bit plastic/shiny rubber-like. That is what I meant by me not understanding why posting images were relevant if you like them and think they are good, and I prefer mine. It is irrelevant, both of us have demonstrated that we can figure out how to type in a prompt.
Here are some examples of what I would consider realism, as the first ones I posted clearly weren't meant to be, but was images, I created making comic styles and for a fantasy board game, except the woman wearing a military uniform that was just for fun:
I do not see any reason to continue this posting images, if you do not think that I have tried AI, despite having created more than 1500+ images, I don't really care what your opinion is about what I asked. As I don't feel that is really what you are interested in anyway.
I don't mean this in a rude way, but I didn't intend for this post to be people posting images. As I mentioned in the OP, I rarely post any images, unless it is examples just like i'm doing now, because you asked, despite finding it completely irrelevant. :)
I don't think you understand me, I don't care if your images are realistic or not, it is completely irrelevant to what I wrote in the OP, I have no clue what people's intentions are when they make an image. If you have worked professionally in 3D, you ought to know that, it should be pretty obvious.
Take this image (I didn't create it, but I do think it is a high quality image, it looks very cool, with nice colors and composition):
Clearly, the person that did, didn't go for realism, does that demonstrate that the person doesn't know how to use AI? or that they suck at it? Of course not, this might be exactly what they were after, and I think it looks cool. Doesn't have to be realistic to be interesting or good.
The images I posted are those I liked, the rest I deleted due to various issues when you work with AI.
It is not a competition, people have different tastes, which is also why it surprises me, that you feel a need to mention that you work professionally and therefore all others' opinions are irrelevant if they don't agree with you.
Even people that have never tried AI, should be able to figure out from the insane amount of images and videos being created that this is probably not educated artists suddenly going crazy with AI, but random people that think it is fun. And therefore also able to have an opinion or just a guess of how this could affect the industry, they might be working with traditional art, but might have seen or experienced AI without having tried it themselves. It is irrelevant to my OP. If you think Im wrong about something I wrote, please let me know, I'm just guessing based on the different AI tools I have tried, what I have read that people have written and the debates I have watched on the topic.
Don't play the victim here.
I try to encourage people to have a civilized talk about AI, and you rush at me like an angry shark.
"Before i even bother to take the time to reply to this post, and its many points, have you actually used AI... Seeing as it is that easy to use and you can turn out images of very high quality in a few minutes, i would like to see who i'm replying to and their understanding of how it work, please show your AI creations here and your understanding of the medium..."
That comment is extremely uncalled for and not only rude but also very arrogant. Yet I comply with what you say and post some images, only for you to continue turning it into some kind of "test" as if you hold the golden standard for what is required for someone to use AI. In which case I bite back and then you get offended by it when I don't agree with your definition of high quality.
Yet, never once do you explain what this "test" is supposed to demonstrate and why it is required for you to "..bother spending your time" to share your expert knowledge, as I assume the reason why the test was so important was that you had a point with it?
To me, the approach you have chosen here is equal to someone asking for help with something about Poser and then someone answering that "they can't be bothered to help them unless the person first demonstrates that they can do the most basic stuff first."
As you say you have 25 years of experience with 3D, so you know how long time it takes to learn. Yet like me, you can't have used AI for a long time, because it obviously hasn't been available for that long and already we can make amazing images as above. I like you have also worked with 3D for a very long time, almost 30 years, given the first program I used was Truespace 1.0, yet mostly as a hobbyist with some professional jobs on the side, but not to the point where I would call myself even semi-professional. My point is that learning 3D takes a very long time, these are complex programs in themselves just to learn how to use effectively, but also they require a lot of let's call it secondary learning, like all the techniques in regards to lighting, materials, camera, animation etc.
My point is, that none of us could make anything even remotely "great" in 3D after just a few months of working with whatever 3d application used. So at least for me, it is very difficult to be impressed by any AI-generated images, not because a lot of them don't look cool, but due to the insane amount of cool images being made, but you also know that there is not 25 years experience behind these images, that is just a fact.
So despite what you might think, I would actually find your view on it interesting. Because both of us are going to face AI-generated 3D very soon, in the same way as those working with 2D are now.
Your response is a little vague. As this thread is about both the good and the bad aspects of AI in the current world explain this. How has a tool that lowers the bar for the creation of visual media and increases the pace of output impacted your workload? If the rise of ai created work is only increasing your workload; then in what way? Before you ask I worked in the animation and film world for over 40 years till my retirement a couple years ago. If you want a list of my film credits; then the by your own methodology please show me something that makes it worth my effort to respond with them. Prove you are worth my, or any others; continued response to your passive aggressive drivel.Well if you think that my....creations are not realistic...and yours are...fair enough...for me it's end of conversation, i am a semi professional 3d arts...have been for 25+ years, and the introduction of AI..has only increased my workload...good luck with it....
parkdalegardener posted at 5:42 AM Thu, 10 August 2023 - #4472289
I'll post a simple ai use that has become almost invaluable for my Poser Blender and other 3D work. AI generated seamless texture tiles. While procedural textures are nothing new, a simple statement or prompt can output dozens of examples of a texture almost instantly. Before you ask about 3D and flat maps realize there are also free online tools that have been available for years that will output normal and the rest of your maps from the inputted texture file. Is this perfect? Maybe not yet, but for an indie dev these types of free use tools without the user being beholden to someone else for their assets can be a game changer. Literally.Agree, this is a very obvious use of AI. I haven't personally used AI that much for making seamless textures yet, as I don't really have any project im working on at the moment, but know that it can be a pretty time-consuming process in general and would definitely explore this further should I start one.
I looked at Substance Alchemist if that is the one you are talking about? But it was about a year ago or something and it looked very cool, but haven't kept up with its development and what state it is in now.
https://www.youtube.com/watch?v=5tFoKjuRno0&ab_channel=AdobeSubstance3D
But I see this more as a tool that makes use of AI, rather than a "one-click" solution or what to say.
As you I do see a huge benefit of AI for especially small developers with limited time and resources, but also big companies obviously, but what are your thoughts when these different AI tools get more integrated into the industry and how it will be affected? Do you think it will be primarily positive or negative?
The tools are already integrated into the industry. Used Photoshop lately? Premiere? Any Adobe app just about actually. Their ai is called Firefly, which for longtime Poser peeps is a little strange. Firefly rendering from Adobe anyone?
No, not Substance. That's not a free use tool. Try this: https://cpetry.github.io/NormalMap-Online/ It's about as simple as it gets.
Up until SDXL "training" was done on 512x512 pixel images. Now it is 1024x1024. Yes, there are caveats to this like any other thing in life. That makes output perfect for UV maps as they expect a square aspect ratio to fill UV space. Clapboard buildings, wooden furniture, stone walls, brick fascia, that type of thing. I would not bother doing skin patches that way as skin is not universal across the object or figure but any other "regularized" surface easily accepts this method. Depending upon the generative tool you use, seamless tiles are as simple as pressing a radio button for seamless output. You do not have to use 4 way seamless. You can do 2 way to vertically or horizontally tile an output.
There is a sea change underway in the industry. There has been for a while. Longer than last Aug when the generative ai "leaks" to the masses happened. Like the death of AM radio and celluloid film; things are moving along. Many will feel that time has passed them by and their methods of expression will be depreciated like the 8-track tape. We see it with people who say they cannot render in Cycles, iRay, Superfly or what have you. That those methods produce inferior output in comparison to the engines they used to use. That the methods involved are too difficult or they cannot do it on their 15 year old core2 duo machine running WinXP. They will continue to use the tools they know to produce the results they expect and enjoy. There is nothing that requires them to change their personal workflow for their particular style or aesthetic.
If they are in a large house or "pro" environment then they had better get with the program. The knowledge requirements are constantly in flux. The bar for entry is considerably higher as more robust knowledge, including the ability to train custom data sets for character and environmental output specific to the property you are currently involved in.
I know about Firefly, but haven't tried it though, so my knowledge is a bit limited about it. I have heard good and not-so-good things about it from people that have tried it. But it is still in beta, so eventually they are going to fix it and improve it. But it looks very promising.
I didn't know about the Normalmap-online so have bookmarked it, as it looks very cool.
I haven't tried SDXL yet, as I don't think my graphic card can handle it, it struggles with SD, but does slightly better if I use ComfyUI, but can't really get around the UI on that one. :) I am considering getting a new graphic card, but are waiting to see if some of them don't drop in price, but also I have my suspicions that they are working on graphic cards that are suited more for AI, given how popular it has become and they all pretty much run on GPU and VRAM. So far I don't have issues just using SD, as I mostly use AI for fun and testing purposes. I still think the biggest issue with AI is the lack of control, even with controlNet or when doing training (Which I have had little success with to be honest :D), I think you are somewhat limited and it is often very difficult to get consistent characters, both when it comes to their look, but also clothing etc. But also the lack of prompting detailed/specific scenes is a huge issue for me and to really embrace it as a valid tool for the stuff I like to create that is more important than the high-resolution testing material that SDXL offers, which is probably also why I haven't even tried giving it a go :).
You could probably use a tileable skin texture, as a sub-texture for dealing with potential seams and to use as a base colour that you can then adjust as you say, but again, most texture tools, at least substance painter which is the one I use, have the option to paint across UV tiles, it's very easy to spot seams now with the current tools. I honestly think it is far more difficult to find high-resolution textures of humans, as they often come with lighting information and other "artefacts" that you don't really want in your texture, so AI could be useful for that as well. But maybe Normalmap-online can fix that or at least improve it?
One idea that I think could be done, that would make real AI texturing, is if you were able to throw a 3D model into a program and either based on a description it could generate the texture. One could imagine a spaceship, where all polygroups come with attached prompts, so, for instance, the outer surface could be "spaceship exterior" and the prompt would simply generate a texture based on whatever was written in that. Tools along this line I could imagine that we will see in the future if they managed to allow for better control.
I agree that there are definitely people that are going to be left behind. But I don't think it is only due to slow hardware or adaptation. Because currently or what is changing is that people that work in the industry come with experience, whether they are self-taught or have a degree, there is still a lot of time invested behind each person. But with AI, you not only get an extremely powerful tool, but it also comes with all the experience that it was trained on and can constantly be improved. The other thing is the extreme productivity that they deliver. One of the fun facts for instance within animation, you often see characters with only 4 fingers, because it saves a lot of time over thousands of frames not having to animate all of them, especially if we are talking 2D hand drawings. But assume that the AI were better at drawing hands than it is, you suddenly have an "animator" that doesn't care about such a thing and what details are needed, which a human workforce might simply not be able to compete with.
So the AI, depending on how it develops, especially if the control gets to a point where you can define characters and it can use these, whether humans can actually compete.
3D-Mobster posted at 10:00 AM Thu, 10 August 2023 - #4472294
Ok; so a lot here to consider. I'll start with I don't use comfy so their node system I cannot comment upon.I know about Firefly, but haven't tried it though, so my knowledge is a bit limited about it. I have heard good and not-so-good things about it from people that have tried it. But it is still in beta, so eventually they are going to fix it and improve it. But it looks very promising.
I didn't know about the Normalmap-online so have bookmarked it, as it looks very cool.
I haven't tried SDXL yet, as I don't think my graphic card can handle it, it struggles with SD, but does slightly better if I use ComfyUI, but can't really get around the UI on that one. :) I am considering getting a new graphic card, but are waiting to see if some of them don't drop in price, but also I have my suspicions that they are working on graphic cards that are suited more for AI, given how popular it has become and they all pretty much run on GPU and VRAM. So far I don't have issues just using SD, as I mostly use AI for fun and testing purposes. I still think the biggest issue with AI is the lack of control, even with controlNet or when doing training (Which I have had little success with to be honest :D), I think you are somewhat limited and it is often very difficult to get consistent characters, both when it comes to their look, but also clothing etc. But also the lack of prompting detailed/specific scenes is a huge issue for me and to really embrace it as a valid tool for the stuff I like to create that is more important than the high-resolution testing material that SDXL offers, which is probably also why I haven't even tried giving it a go :).
You could probably use a tileable skin texture, as a sub-texture for dealing with potential seams and to use as a base colour that you can then adjust as you say, but again, most texture tools, at least substance painter which is the one I use, have the option to paint across UV tiles, it's very easy to spot seams now with the current tools. I honestly think it is far more difficult to find high-resolution textures of humans, as they often come with lighting information and other "artefacts" that you don't really want in your texture, so AI could be useful for that as well. But maybe Normalmap-online can fix that or at least improve it?
One idea that I think could be done, that would make real AI texturing, is if you were able to throw a 3D model into a program and either based on a description it could generate the texture. One could imagine a spaceship, where all polygroups come with attached prompts, so, for instance, the outer surface could be "spaceship exterior" and the prompt would simply generate a texture based on whatever was written in that. Tools along this line I could imagine that we will see in the future if they managed to allow for better control.
I agree that there are definitely people that are going to be left behind. But I don't think it is only due to slow hardware or adaptation. Because currently or what is changing is that people that work in the industry come with experience, whether they are self-taught or have a degree, there is still a lot of time invested behind each person. But with AI, you not only get an extremely powerful tool, but it also comes with all the experience that it was trained on and can constantly be improved. The other thing is the extreme productivity that they deliver. One of the fun facts for instance within animation, you often see characters with only 4 fingers, because it saves a lot of time over thousands of frames not having to animate all of them, especially if we are talking 2D hand drawings. But assume that the AI were better at drawing hands than it is, you suddenly have an "animator" that doesn't care about such a thing and what details are needed, which a human workforce might simply not be able to compete with.
So the AI, depending on how it develops, especially if the control gets to a point where you can define characters and it can use these, whether humans can actually compete.
I have two systems here. A 20 series 8 gig and a 30 series 10 gig. There is nothing the 8 gig card cannot touch, including XL generations if I run the refiner weights as a second step. It can be used to train 1.x and 2.x LoRas, TIs, and the like. I cannot train XL on the 8 gig card.
The 10 gig card can load the whole XL pipeline as a single text to image operation the way it supposed to be used according to the research papers.
Adobe's in/out painting could be a bit better it is true if you are comparing it's output to that of other online generators who use custom blends and trainings based on 1.x. Adobe, at least; is trying to separate itself from the ethics of current data sets used in training. The censorship and training upon user created imagery opens a different debate.
Controlnet is a gift from the gods when used in conjunction with any 3D application and a miracle machine even on 2d images. It too can output a normal or depth map without using an outside application like normal map online and implement them during generation. Segmentation is what selective matte making for free looks like. It is the method of doing your spaceship. Run the segmentation net and it's all point and click. For instance you can click on a tailfin and the ship body they become the same material in that specific situation. Perfect mattes that you can add and subtract into one or more material regions. Not prime time yet for 3d models until someone gets better backside generation and I don't see that happening soon come.
Pose estimation gives fantastic results when used with a simple doll posed in your 3d software of choice. Controlnet can grab the pose from your doll and apply it to whatever figure type you may have in mind as ai generated output. For instance pose Andy in Poser or any other figure in your software of choice and kick out an image. Put it in controlnet while asking for a bunny dressed as a viking warrior and you get a bunny dressed as a viking warrior in the pose controlnet extracted from the Andy image. Works for facial expressions and hands too. I am waiting for similar control weights for SDXL.
The VRAM issue may be driver specific. Newer graphic drivers and every new "game ready" driver offload VRAM to system RAM due to the need to load a large number of 4k+ texture files and video information at the load of newer games. Offloading VRAM for texture files is no big deal. Offloading VRAM for ML tasks decreases the ability to actually run some tasks or to perform them in a "reasonable" amount of time. Older studio drivers did not do this.
The amount of control one gets is actually very good. Again; it is far from perfect, but it is very good. That is where the need for artistic types remain alongside the technicians. In a large professional environment the ability to rapidly output product does not equate to a job well done. One must realize that while rapid, ai is not exactly smart. If I tell my art department that I need product placement in a particular setting ai cannot do it. It can put "a" product into a scene but not "your" product into a scene if all you are doing is typing a phrase into a prompt box.
Send the product to the art department and have the photogs snap a pic or ten. Those pics are uploaded to the tech crew as they are being taken real time. Ten minutes and you have training data. A decent 40 series card with a whack of VRAM and you can output a LoRa for your product in as little as a half hour. At this point you can get a reasonable output image of "your" product placement in the setting required as quickly as you can type in a prompt. So an hour or so of prep work and you can instantly have "your" product placed wherever. The question remains though. Does typing your phrase into the prompt box make you a good image now? Probably. Will it sell your product or put forward your idea? Probably not. That is where the "artistic" type people come in.
I see the need to have reasonable tech folk that can train the weights (models) and ethically acquire the training data, from in-house or elsewhere to be one part of the team; while one still needs artistic folk to get the most out of the tools to provide the return companies need to survive. A thousand monkeys typing on a thousand typewriters might produce a reasonable Netflix pilot in the long run, but you get a lot of drivel while you wait. In a professional environment that doesn't work. You hire trained monkeys. It's time to become a trained monkey if this is the type of work one does.
What we consider animation has changes a lot. I know. I trained first as a photographer and then as a film maker / special effect artist in the mid 70s. Everything I learned in university is ancient history. I changed and became an animator and watched auto tweening programs like Poser force me to adapt or or die. Well no problem. I'll learn to 3D model. That will keep me relevant I thought. And that is the point. Adapt or die.
If I was a mechanic I would have become obsolete had I not learned about electronic ignition as the carburettor has been replaced by it, and now electric cars will make ICE a thing of the past. Plug in a car and the machine tells you what is wrong with the engine. It then tells you all the steps in order to fix the issue. Button pusher does just that and the problem persists. Now I need my mechanic. The mechanic is the artist that can finess the machine.
My issue is that I only have a 6gb and from what I could figure out it has some issues with "half precision" or something, which makes it less good for AI (Not going to pretend that I know what it is) But even without this issue 6gb is on the limit already :)
I didn't want to make it sound as if controlNet is not useful, it is very useful, but I still think in its current state, probably due to how the AI-generated technology works that it has its limitations as you say it isn't exactly smart. And btw I think I read that it was released for SDXL the other day, but might be mistaken as I only read the headline :D
The issue with AI is consistency in my opinion, but what you can do is use img2img combined with PS mockups, it will not completely solve the issue of consistency, but if you keep repeating this, you can to some degree control the image, also there is support for transparency as far as I know to control which areas should be generated. But obviously, it ain't really the optimal workflow.
I agree with your product example, but im not so sure that it will stay like that as the AI tools improve, because even when doing let's say product presentation there are some techniques that work better than others obviously depending on the product. But I don't see any reason why an AI couldn't learn that, especially if you got some AI image generation tool integrated with ChatGPT or whatever.
For instance, I just did a quick example using Chatgpt, to make up some cooking book and then told it to use the AIDA model for writing a description to sell the book:
Discover "Wild Feasts: Cooking Beyond Campfires"!
Are you ready for a culinary adventure that takes you beyond the ordinary? Dive into the pages of "Wild Feasts" and explore the thrill of cooking in the great outdoors. Imagine creating mouthwatering dishes surrounded by nature's beauty and flavors. Delve into the art of foraging, master shelter-cooking techniques, and infuse your meals with the essence of the wild. From angler-inspired delights to backcountry gourmet wonders, this book is your guide to extraordinary outdoor cooking. Are you prepared to unleash your inner wild chef and turn every meal into a memorable feast under the open sky?
I didn't spend a whole lot of time on this, but I think it illustrates the idea well enough that one could imagine that it could also be done with a product image, as long as the technique for what works is available. Let's say you had to make a commercial for travelling to some nice sunny place, then you wouldn't make an image of it raining or put some unattractive and angry people in the image. So my point is, that if the AI (Chatgpt) knows this and my guess is that it probably won't take long before it can be fully integrated with an AI image generation tool, I think ChatGPT can already create prompts if im not mistaken. Then the margin for why you would spend 1000s of dollars hiring someone to tell you this might be somewhat blurred, compared to just telling the AI to generate an amazing image using well-known techniques.
I agree if you just run with the AI image generation alone then you will as you say, probably get something decent out, but it might not sell the product as well as one might hope. The question is obviously how all this AI is going to work in the future, we know that people are very creative when it comes to making use of these tools and new technologies in unexpected ways.
Also want to stress, that I don't think that AI will replace all humans, so when I talk about the potential impact Im simply speaking about whether it will reduce the number of people required. Just like I don't think that AI will replace all journalists, but there are definitely some that will be replaced I think, simply because they become more effective.
I don't think it is as much about adapting, as you might. Because I agree with your example about the mechanic in the scenario you use. But I think it is an incorrect scenario of what is going on. Because it should be, the mechanic repairing cars and then some robot (AI) thingy comes along and can repair the same cars day and night and much faster, and maybe 5 cars at a time. The AI image generation tools, in broad terms, don't really do anything that humans couldn't do, they just do it insanely much faster and in a way that makes it extremely easy to do. It could probably be compared to a person driving their car into a garage and then pressing a button on the computer saying "Fix my car". So even if the AI can't fix all things that a mechanic could, it might be able to fix let's say 70% of the things, which would still mean that mechanics would lose that percentage.
"The good and bad things about AI generated images?"
Good:
1) Royalty-free, public domain, anyone can re-use raw generated image. Open source software.
2) Quality is increasing every day (SDXL etc), and vastly better that a year ago.
3) Images can be generated for free.
4) Imaged can generatively iterated. Get something near to what you want, then work it towards the ideal.
5) Happy accidents, within the framework of lots of preset styles.
6) Create things that are wildly imaginative / surreal / dream-like.
7) Can 'train' an AI model on a relatively small sub-type of images.
8) Increasingly controllable. Increasingly possible to run locally.
9) Take a 'rough' figure render from Poser or DAZ, and do amazing things with it.
10. It's naughty (in most people's eyes). We could use a bit more 1960/70s naughtiness in this conformist age.
Bad:
1) Gloopy, loopy, 'spider-webbed' images still occur, full of problems.
2) Images usually need a lot more fixing in Photoshop than you'd expect at first glance.
3) Over-coloured. It's often difficult to get a lower-key coloring scheme.
4) Used by the clueless, who have no idea how to edit a set of pictures down to two or three of the best ones.
5) Used by the clueless, who have no idea how to fix the problems in Photoshop.
6) Still likely to mangle and distort hands, feet, vehicles with wheels, in-scene signs and logos.
7) Pay-to-play, unless you can find free or local desktop PC options.
8) Lots of time wasted, getting it to do what you want it to. it's still a bit of game.
9) No streamlined start-to-finish production software, yet. No "Comic Life 4 with built-in AI", for instance.
10) May cause intense whining, in online forums.
Learn the Secrets of Poser 11 and Line-art Filters.
11) may get you unfriended and blocked on social media
not sure if this belongs in the good or bad category ;-)
Yeah, social media can be rough, try chatting about movies and people will go nuts if they disagree with you :D
Anyway I found this article, which I think is interesting.
-----
Freelance illustrator Amber Yu used to make 3,000 to 7,000 yuan ($430 to $1,000) for every video game poster she drew. Making the promotional posters, published on social media to attract players and introduce new features, was skill-intensive and time-consuming. Once, she spent an entire week completing one illustration of a woman dressed in traditional Chinese attire performing a lion dance — first making a sketch on Adobe Photoshop, then carefully refining the outlines and adding colors.
But since February, these job opportunities have vanished, Yu told Rest of World. Gaming companies, equipped with AI image generators, can create a similar illustration in seconds. Yu said they now simply offer to commission her for small fixes, like tweaking the lighting and skewed body parts, for a tenth of her original rate.
...Chinese video game companies, from tech giants like Tencent to indie game developers, have begun using these programs to design and create video game characters, backdrops, and promotional materials.
(Tencent might not be well known by a lot of people, but it is a massive company owning I don't know how many gaming studios. And have a net worth of 402 billion dollars.)
“AI is developing at a speed way beyond our imagination,” Xu Yingying, illustrator at an independent game art studio in Chongqing, told Rest of World. Xu’s studio produces designs for major game developers in China. Five of the studio’s 15 illustrators who specialize in character design were laid off this year, and Xu believes the adoption of AI image generators was partly to blame. “Two people could potentially do the work that used to be done by 10,” she said.
AI-generated art was so skilled that some illustrators talked about giving up drawing altogether. “Our way of making a living is suddenly destroyed,” said a game artist in Guangdong, speaking on condition of anonymity for fear of being identified by her employer, to Rest of World. Yu, the freelance illustrator, said it was “despicable” that algorithms — trained on vast datasets that took humans decades to produce — were on the verge of replacing the artists themselves. Still, Yu plans to train AI programs with her own drawings to improve her productivity. “If I’m a top-notch artist, I might be able to boycott [them]. But I have to eat.”
-----
And still, we are only in the early ages of AI, so it doesn't look too good for traditional artists I think, unfortunately. Im wondering if there is anyone out there that has a solution to this or whether they just hope that it will somehow solve itself or just accept that this is how it is going to be?
it just happened in our lifetime
past technologies caused many to literally starve to death we conveniently forget about
as we enjoy our thousand thread count cotton sheets
nobody is taking about the Newsagents either
one can still enjoy doing art, the printing press reduced but didn't kill calligraphy
hobbies don't cease to exist, many garden, knit and sew etc
not many make a living breeding horses either, it's a costly hobby unless for the cruel short life of racing and pet food industry
not dismissing the problem just stating cold hard sad facts but as always messengers get shot
The way AI-generated images are evolving and providing opportunities for indie developers and artists is truly remarkable. On the flip side, I understand the concerns about creativity and job opportunities for traditional artists, but I believe that artists will be the ones to control the artificial intellect, materializing their thoughts through prompts that will generate.
I have a few friends using Mid Journey AI to generate some stunning arts that they post on Instagram and get a lot of likes and followers with them. I call them a kind of "AI Artists." I've used it for various purposes, from creating logos to site images and wallpapers. If you want to explore it, you can download Mid Journey from here and try it. It's really handy, and I've found that it opens up possibilities rather than stifles creativity.
Frankly I treat AI like the toy it is. I don't consider myself an artist because of what it produces. I use it because it's fun. People who demand recognition for the works done by an AI suck the fun and life out of it! As exhibited by DarkGlass when he comes in demanding you post images in a thread about discussing AI as if you can't have an intelligent discussion unless you can post pretty pictures.
yeah I refuse to get into an argument about wanting compliments for using an AI image generator trained on a database
I still enjoy the results and will say so
I probably will just share the occasional image in said thread and refrain from discussion altogether
not taking "sides" BTW the only argument is whether one should be using AI at all and obviously I am in the group OK with it
SteveCampsOut posted at 5:30 PM Sat, 19 August 2023 - #4473013
I would probably agree with this at least to some extent. What impresses me is the AI tool, not the person typing the prompt. But when AI gets to the point where you can use it to tell a story, which is what really interests me, I would love to be able to make comics using it. But unfortunately, I don't think it is really there at the moment. But then again, it wouldn't be the art itself that would impress me, it would be the story told.Frankly I treat AI like the toy it is. I don't consider myself an artist because of what it produces. I use it because it's fun. People who demand recognition for the works done by an AI suck the fun and life out of it! As exhibited by DarkGlass when he comes in demanding you post images in a thread about discussing AI as if you can't have an intelligent discussion unless you can post pretty pictures.
But I think time will tell when these tools get more developed. But at the moment at least it seems to be mostly people making "random" images and videos and having fun with that.
The chance that these AI tools keep pushing the boundary might eventually make it so none of what is created is really impressive, regardless of what it is and then it will be more about entertainment rather than art. At some point when the first few 100% AI-generated movies have been made and people are no longer impressed by them, then what we want are good stories and it will simply be expected that these movies/comics, whatever look amazing by default.
Well, agree or disagree. AI is a tool that can take the written word description and run with it to create every sort of image from crap to the creation of the universe, if the writer is elegant enough and the AI has enough data to extrapolate such a vision. Is there a talent to word smithing? Certainly. AI art isn't totally the creation of the AI. It needs the inspiration of the words used by the author along with the data stored in its database to come up with something worth viewing. Am I going to get puffed up with pride over the output of an AI? Not really. Will I enjoy what it produces? Of course, I will if it's aesthetically pleasing to my eye.
Privacy Notice
This site uses cookies to deliver the best experience. Our own cookies make user accounts and other features possible. Third-party cookies are used to display relevant ads and to analyze how Renderosity is used. By using our site, you acknowledge that you have read and understood our Terms of Service, including our Cookie Policy and our Privacy Policy.
I think it comes as no surprise that there are a lot of "heated" debates going on between traditional artists and those embracing AI-generated art. Personally, I haven't experienced any negative feedback, but then again I don't really post a lot of images either on forums or even here with the purpose of getting feedback, but more as examples.
But looking on Reddit, there are often posts of people that feel harassed by one or the other side. However this is not really what this post is about, but rather what benefits or negative things you experience or feel AI-generated images have at the moment and in the future once it gets more developed.
As everyone should know by now, AI-generated images are also venturing into the 3d realm fast and it will probably not take that many years before it is a valid tool in this area as well.
So what positive things do you see with it and what concerns do you have?
(Let's try to respect each other's views :D)
My own view is that im neither for nor against it, I enjoy all new tools and focus on the end product, how one gets there is of little interest to me.
Starting with the benefits of AI.
Ease of use and speed, that you can turn out images of very high quality in a few minutes is pretty damn amazing. As an indie developer, whether that is you wanting to make games, both computer games or physical games having access to close to an infinite amount of assets is absolutely a game changer. But pretty much in any case where the budget is limited and having to spend a lot of money on artists is simply not possible for a lot of people.
That it is open source is just absolutely amazing, again for small indie studios/individuals this makes it so much easier for them to realize their dreams.
Looking at 3d applications, I think it would probably be possible over time to optimize and improve the quality as well, for instance as many know, hair in 3d is an absolute pain and often look very bad, simply because it doesn't seem to flow naturally, and even doing cloth animations can in some cases be very time-consuming. One could imagine that by mixing AI into the workflow, it might be possible to add natural-looking hair and cloth as an after-effect. One could also imagine it being integrated into the animation of characters so rather than having to animate all by hand one could use a prompt. Obviously, these things might be somewhat into the future, but at least the possibilities are there.
Looking at the bad things, the most obvious to me at least is that pretty much anyone at the moment can generate high-quality images with very little effort and it is going to harm traditional artists' job possibilities. It might even discourage younger people from even entering the field in the first place, why study for years to become an artist if you can just type it in an AI tool? Even the best artists out there, can't keep up in production with a computer that is constantly improving and is more productive than they are.
Also, a lot of people are taking advantage of these new AI to exploit or "scam" people that have little knowledge of it, meaning that people might risk paying far more for something that they think is an amazing piece of work, but that is in fact generated very quickly using AI.
I don't think traditional artists are in any particular danger and AI have little impact on them. But especially people making a living from digital art or work, whether that is making commercials, photographs, 2D illustrations or even 3D illustrations are facing some very dark times in my opinion.
Some open questions, could be whether creativity could be hurt over time or not? One argument could be that given that as more and more productions start making use of AI, some of the human creativity is lost as it will be more of a "factory" production of just spitting out new things. People won't really be impressed by it for very long. Alternatively, one could make the argument that with AI, there are so many more possibilities, as you don't really have to limit yourself, the AI will spit out whatever you need regardless of how complicated it is.
In the somewhat distant future (5-10 years), my guess is that the whole industry is completely different from what it is now and if people even do "traditional" stuff as they do now, I have my doubts about, simply because the market for it might not be viable, it is too expensive to hire these people compared to those working with AI, but it obviously depends on where this technology is going, but one thing is for sure it is going extremely fast :)
What are your thoughts on AI both now and in the future?