There's a great deal of pushback against AI art from the wider online art community at the moment, a lot of which is motivated by a sense of unfairness: if you're not going to put in the time and effort, why do you deserve to create such high equality imagery?
(I do not share this opinion myself, but it's something I've seen a lot)
This is another great counter-example showing how much work it takes to get the best, deliberate results out of these tools.
This is not something I've seen once in any sort of criticism of "AI art", and elsewhere in the internet I'm largely in a anti-ai-art bubble.
Most legitimate pushback I've seen has been more on the non-consensual training of models. Many artists don't want their work to be sucked up into the "AI Borg Model" and then regurgitated by someone else, removing the artists consent, credit, and compensation.
> "...and if we allow this, those years were wasted and I'm out of a job, along with millions of others"
Given that the quoted part is highly likely to happen, and not just in art, I hope we're no longer considering "luddism" to be a pejorative.
Technical progress does really require adaption sometimes. We don't criticize luddites because we think they were wrong about change being a real thing.
I wouldn't consider it a pejorative either because I don't see artists mobbing datacenters to smash A100 clusters with pitchforks.
I see it as a slow transition though; there's still plenty a human artist can offer over any current model even with a carefully curated prompt. But yes, eventually the whole industry will die down. Especially seeing as models can now generate sounds, 3d models, textures, natural placement of objects on a map, etc. Like everything we've invented a tool to help us do things faster and it will displace people. Tough to say whether it's right or wrong but it's what we've done all through history; move on from one technology to the next. I wonder if traditional artists complained/fretted when digital art/tablets were getting big, hmm.
Using stock art is just further appropriation, which is silly considering the intent and licensing of stock artwork is clearly intended by all parties to turn works into commodities for commercial exploitation.
The old ways are best, the new ways are bad and take away the soul from the creation process and resulting works. Also unconvincing considering that most of the people saying that are using radically different, digitized, heavily time optimized art workflows compared to norm of the industry even 30 years ago.
Not that I don't see the problems, the potential for job losses due to the optimizations to workflow requiring less work and therefore less workers is an actual risk, but one that happens regardless of copyright enforcement of AI models. The problems commercialized AI art workflows cause may even be exacerbated by enforcement of copyright on training data by handing a monopoly of all higher quality generative AI models into already entrenched multinational intellectual property rightsholders hands. I think a lot of artists forget that copyright isn't as much for them as it's for the Disneys of the world.
In order to copy style from other artists reliably - you have to make a LoRA yourself. That involves a lot of manual work and it can't really be automated if you want good results.
Artists can opt out of future SD base models (which doesn't matter), but they can't opt out of someone making a LoRA of their work (which actually works).
Is this part actually coming from artists? What’s the suggested amount(be it upper quadrillion dollars per second or $0.25/use)?
I think compensation as a condition is only assumed implied, that financial gains are artists’ motives and they actually live off that income. Rather, I see a lot of vocal oppositions to AI image generators that aren’t drawing for profit at all.
So, is the money going to solve it, or is it a wrong assumption, or is it that it will have to be settled by lump sums?
Yes. The group of artists that are suing Stability AI and Midjourney are calling for consent, credit, and compensation. https://stablediffusionlitigation.com/
> Since then, we’ve heard from people all over the world—especially writers, artists, programmers, and other creators—who are concerned about AI systems being trained on vast amounts of copyrighted work with no consent, no credit, and no compensation.
I think the details of credit and compenstation aren't as important, because once you require consent, artists can decide whether they're happy with the compensation model and choose to give consent (or not) based on that.
Look at the pushback to Adobe’s model.
“Non consent of model input” is just a tool they’re using in the hopes of destroying the tech. Plenty of companies have datasets of these same people’s work where the T&C permits training.
The narrative will switch once you can no longer use the “stealing/consent” argument. They won’t suddenly become fine with this tech just because the dataset consented.
I work in the creative fields so my work will be impacted by this but I realize it’s pointless to fight it.
I've actually seen this a lot.
In my view, it's not coming from professional artists working in the field. Their concern is more that people are ripping off their style, or that AI is making their efforts unnecessary (e.g. lots of people who made a living by copying the style of particular anime & cartoons for fans, no longer have a purpose since AI can do that given enough source material).
Non-professional artists, on the other hand, are still learning and have put a lot of time into their craft and it hasn't paid off yet. They seem to be annoyed that other people are getting results (via AI), without actually having to learn the mechanics of art.
AI basically lets your generic art history major produce lots and lots of pieces, because they can describe artwork well enough and know where to find good samples for the AI. The only thing stopping them was mere mechanical inability, not knowledge of the art space.
Some modern AI art workflows often require more effort than actually illustrating using conventional media. And this blog post doesn't even get into ControlNet.
Indeed. Another criticism that I can definitely somewhat see the idea behind, is that the barrier to entry is very different from for example drawing. To draw, you need a pen and a paper, and you can basically start. To start with Stable Diffusion et al, you need either A) paid access to a service, B) money to purchase moderately powerful hardware or C) money to rent moderately powerful hardware. One way or another, if you want to practice AI generated art, you need more money than what a pen and paper cost.
There are plenty of free online tools for using all kinds of AI image generation techniques, and they don't require powerful hardware, just something that can browse websites or run Discord.
It’s like with dreams. They can be terribly intricate and detailed, but ask me to draw something creative and I’m out.
Our DM, being someone who has released creative works, was reticent and less gung-ho on AI for awhile until he decided to start playing with the AI tools (Midjourney in his case) for a new custom campaign he's running. He's suddenly able to develop novel NPC tokens for every important character we meet, the monsters are high resolution and the convenience of using Midjourney in Discord (which he already uses to coordinate our online D&D games) has been a huge boon and enhancement to how much fun and immersive our games are. A year ago this would have cost literally tens of thousands of dollars. He's a published fantasy author so prompting aka describing a scene comes naturally to him. It's been a lot of fun seeing what he's coming up with.
I'm really loving the spark of creativity I've been finding in myself where the turnaround on the old tools was too long for me to not get frustrated and give up, and to see it amongst my friends, even the ones who were initially skeptical.
With a super-cheap T4 GPU (free in Google Colab), PyTorch 2.0, and the latest diffusers package, you can now generate batches of 9-10 images in about the same time it took to 4 images when Stable Diffusion was first released. This drastically speeds up the cherry-picking and iteration processes: https://pytorch.org/blog/accelerated-diffusers-pt-20/
Google Cloud Platform also now has preview access to L4 GPUs, which are 1.5x the cost of a T4 GPU but 3x throughput for Stable diffusion workflows (maybe more given the PyTorch 2.0 improvements for newer architectures), although I haven't tested it: https://cloud.google.com/blog/products/compute/introducing-g...
A 4GB NVidia GPU (sufficient to run Stable Diffusion with the A1111 UI) is hardly “moderately powerful hardware”, and, beyond that, Stable Horde (AI Horde) exists.
OTOH, a computer and internet connection are more expensive than a pencil, even if nearly ubiquitous.
Thqt got me thinking. I agree, but from another perspective: the skillset is different. Traditionally, the approach to art was very bottom-up. Start with a pen and basic contouring techniques. Understanding more advanced techniques require a lot of work (perspective, shadows, etc).
"AI" art generally does away with basic techniques. The emphasis is more on composing, styling. A top-down approach. "AI" artists may be able to iterate quicker by seeing "almost-finished" versions quickly (though a skilled artist can most likely imagine their work pretty well).
But most of all, the tools and required skills are very different. You don't need to know a lot about machine learning, but it certainly helps. Probably pretty far from the skillset of most current artists. And people generally fear what they don't understand. And if I was an artist, I'd be at least a bit concerned about (i) it undercutting the value of my art, (ii) having to learn this alien way of doing things to remain competitive (by way of selection, artists probably enjoy their current tools.
Anyway, I imagine photography was similarly upsetting in a lot of ways. It also didn't happen overnight. I also suspect we are going to see similar improvements to output quality as in early days of photography.
Another similarity is with digital music (and recording/remixing before that). I wonder if we're going to see new genres emerge as a result (the equivalent of techno/electro).
This complain in particular strikes me as someone who enjoyed the process more than the result. Very specialized crafting skills, now made, not useless, but not required to obtain a similar result. And, if reducing things to a market, competing with a very orthogonal set of skills.
"3D Stable Diffusion" doesn't exist yet, but some projects look promising.
None of this is true, you can easily use a colab and load the models on a drive, I have always done it this way and it works perfectly.
Also if you just need to start, there are plenty of free interface online to use, colab if actually you want to dive in.
Even if they did have a more complex workflow most of them are still based on copyrighted training data, so there will be many lawsuits.
people that want to hone a niche discipline for themselves still can do that. just be honest about doing it for yourself.
People will be arguing about whether or not art made with AI is art, and artists will just be using it or not. I remember an interview about electronic music where Bjork addressed concerns that if you use a computer to make music, it has no soul, and she said if the person using the machine to make the music puts soul into it, it will have a soul.
I remember David Bowie in the mid 90s saying if he was young in that decade he might not have been a musician, because in the 60s being a musician seemed subversive and at the time of the interview the internet was carrying the flag of subversion.
Anyway, it's interesting to watch these conversations. I'd never claim to know what art is or try to tell someone, but it seems to me that already because of the controversy artists are drawn to AI and further exciting the conversation. Commercial artists seem the most threatened; animators, designers, etc. I understand why, but I don't think arguing that AI isn't "art" is going to help their cause any more than protesting digital painting wasn't art, electronic music wasn't art, and much earlier that photography wasn't art.
All the time these conversations are happening, the art's getting made and we're barreling towards the next 'not art' movement.
Except all those artists’ art being used without their consent to train these models that subvert the exclusivity of their style, or obviate their work altogether. It ingests their literal effort and eliminates other humans’ need to put forth a similar level of effort.
I’m all for cool new tools, but this is very much like the invention of digital sampling, and models should be required to “clear” all the works that they “sample” for training.
As someone just brought up previously, though, there are other AI art models that don't use other people's work without permission.
For example, the new adobe ai stuff is done this way. And yet people will refuse to concede the argument at that point, and still think it is bad.
Then why don’t they illustrate it instead, and save themselves some time?
I disagree - I think that AI generative art is an easy case of copyright infrigement and an easy win for a bunch of good lawyers.
That's because you can't find an artist for a generated picture other than the ones in the training set. If you can't find a new artist, then the picture belongs to the old ones, so to speak. I really dont see what's difficult with that case. I think the internet assume a bit to quickly it's a difficult question and a grey area when maybe it just isnt.
It's noteworthy that Adobe did things differently than the others and the way they did things goes in the direction im describing here. Maybe it's just confirmation bias.
“you can't find an artist for a generated picture other than the ones in the training set. If you can't find a new artist, then the picture belongs to the old ones, so to speak”
I don’t think that’s valid on its own as a way to completely discount considering how directly it’s using the data. As an extreme example, what if I averaged all the colours in the training data together and used the resulting colour as the seed for some randomly generated fractal or something? You could apply the same arguments - there is no artist except the original ones in the training set - and yet I don’t think any reasonable person would say that the result obviously belongs to every single copyright owner from the training set
Artists and designers have furthered their careers (and gained notoriety) by 'ripping off' others since the dawn of time. This used to require technical artistic ability; now less so. The barrier to entry is.... not necessarily lower now, but different.
Normally - outside the specific context of AI generated art -, there is not a relation "work¹ → past author" , but "work → large amount of past experience". (¹"work": in the sense of product, output etc.)
If the generative AI is badly programmed, it will copy the style of Smith. If properly programmed, it will "take into account" the style of Smith. There is a difference between learning and copying. Your tool can copy - if you do it properly, it can learn.
All artists work in a way "post consideration of a finite number of past artists in their training set".
There's already a line, what does it matter who produced the work? Just judge everyone using the same criteria.
Even if you don't find the moral argument convincing (and you should), legally protecting creators will encourage creativity, which is the whole point of copyright, while allowing anyone with a laptop to coopt their work will discourage it.
But, yes, if Midjourney trained on copyrighted works, anything produced by it is in my opinion IP theft.
It doesn't belong to the "old ones", it is at best a derivative work. And even writing a prompt, as trivial as it might seem, makes you an artist. There are modern artists exposing a random shit as art, and you may or may not like it, but they are legally artists, and it is their work.
The question is about fair use. That is, are you allowed to use pictures in the dataset without permission. It is a tricky question. On one extreme, you won't be able to do anything withing infringing some kind of copyright. Used the same color as I did? I will sue you. On the other extreme, you essentially abolish intellectual property. Copying another artist style in your own work is usually fair use, and that's essentially what generative AI do, so I guess that's how it will go, but it will most likely depends on how judges and legislators see the thing, and different countries probably will have different ideas.
> That’s because you can’t find an artist for a generated picture other than the ones in the training set.
First, that’s clearly not true when you are using ControlNet with the input being human generated, or even img2img with a human generated image, but second and more importantly…
> If you can’t find a new artist, then the picture belongs to the old ones, so to speak.
That’s not how copyright law works. The clearest example (not particularly germane to the computer generation case, but clearly illustrative of the fact that “can’t find another artist” is far from dispositive) is Fair Use noncommercial timeshifting of an existing work: it is extremely clear there is no artist but that of the original work, and yet it is not copyright infringement.
> I really dont see what’s difficult with that case.
You’ve basically invented a rule of thumb out of thin air, and observed that it would not be a difficult case if your rule of thumb was how copyright law works.
Your observation seems correct to that extent, the problem is that it has nothing to do with copyright law.
> I think the internet assume a bit to quickly it’s a difficult question and a grey area when maybe it just isn’t.
IP law experts have said that the Fair Use argument is hard to resolve.
Assuming the lawsuits currently ongoing aren’t settled, we’ll know when they are resolved what the answer is.
Agreed. An AI model trained on an artist's work without permission is IP infringement and this should be widely understood. Unfortunately, because the technology is new people do not understand this. When Photoshop was new, there was a similar misunderstanding. People could take an artist's work, run it through Photoshop, and then not compensate the artist. It took some time for that to sort out.
It’s a static mapping, surely it should be possible, you’d think, but NN frameworks aren’t designed that way. That is blocking it from happening(and also allowing “AI is just learning, human is same” fallacy)
We have some countries where it is explicitly legal to train AI models on copyrighted data without consent, and precedent in the US that makes this a plausible outcome there as well.
Could you explain what portion of copyright law you believe would cover this argument? I'm not a lawyer, but have a passing familiarity with US copyright law, and in it, at least, I do not know of anything that would support the idea you're proposing here. How would you even assign copyright to the "old" artists? How are you going to determine what percentage of any given generation was influenced by artists X, Y, Z?
Nonetheless, it would be odd and a weak argument to point criticism towards not spending adequate «time and effort» (as if it made sense to renounce tools and work through unnecessary fatigue and wasting time). More proper criticism could be in the direction of "you can produce pleasing graphics but you may not know what you are doing".
This said, I'd say that Stable Diffusion is a milestone of a tool, incredible to have (though difficult to control). I'd also say that the results of the latest Midjourney (though quite resistant to control) are at "speechless" level. (Noting in case some had not yet checked.)
I don't get this. If one "can produce pleasing graphics," how does that not equal knowing what they're doing? I only see this as being true in the sense of "Sure, you can get places quickly in a car, but you don't really know how it works."
The goal may not be to produce something pleasant. The artist will want some degree of artistic value; the communicator will want a high degree of effectiveness etc. The professional will implicitly decide a large number of details, in a supposedly consistent idea of the full aim. The non professional armed with some generative AI tool may on the contrary leave a lot to randomness - and obtain a "pleasant" result, but without real involvement, without being the real author nor, largely, the actual director.
What are you suggesting, that the author lied in the blog post and actually worked backwards, post hoc? Seems incredibly unlikely based on the details they wrote.
"Pleasant" is not necessarily the goal. The goal may be "artistic", which implies some inner logic that the artist intends to express; the goal may be "communicative", which implies that the creator will focus on the effectiveness of the thought-structure presentation.
There is an amount of knowledge that is in the professional that may not be codified for explicit fruition in the machine. This may for example suggest that the knowledgeable author will want a very high degree of control on the output, as the details are involved in his intention - so, he will not leave much space for random randomness.
The z-movies director will shout "cut, good, keep this, next" very frequently; Kubrick will shoot scenes in dozens of takes, even over a hundred. Apprentices in music will explore; Hollis directed authentic professional musicians for very long sessions looking for specific qualities in results, and could delete entire parts for being imperfect or alien, while even excessively nice.
Perfectionist authors are demanding in all sides of creation. (And delegation, as said above, not just detail, will not be left to randomness.)
This isn’t high quality imagery. Don’t get me wrong, the tech is cool and I love the work that’s went into making this picture. But this isn’t something I would ever hang on my wall. There’s probably a market for it, but I get the strong impression it’s the “live, laugh, love” market. The people that buy pictures for their wall in the supermarket. The kind of people who pay individual artists money to paint bespoke images of their pet are not going to frame AI art. I don’t think the artists need to worry.
In-ai-aided art, like manually developed film, will trend towards a niche.
Well yeah but that doesn't change the OP commenter's point that it takes a lot of work to get high quality art still.
> I don’t think the artists need to worry.
I disagree here but only on the basis of what type of art it is. Stock art/photography, and a lot of media designwork is likely at risk because we can now create "good enough" art at the click of a button for almost no cost. I agree that the "hang on the wall level good" artists aren't at risk just yet, but between the more filler-art and the uh
Well "anime/furry" commissioners are definitely at risk right now for anything except the highest quality artists, and there is a MASSIVE community behind this - in fact they have done a lot of the innovation for StableDiffusion including optimizations/A1111 webui, and have trained many custom models for their art, already had pretagged datasets of 10k's of images....
What the model can do best is to convert a basic contour sketch into an elaborate rendering, which means that it's the top end of the market that was doing those renderings that's most at risk.
I’ve done pictures of my wife in the style of other photographers, Soviet-style propaganda posters, 50s pinups, Alphonse Mucha, and much more.
I’m a professional photographer and have tons of great pictures of our dog - the kind of stuff people pay for. My wife’s lock screen on her phone is something I generated instead.
It's interesting to ask people who are concerned about the training data what they think of Adobe Firefly, which is strictly trained on correctly licensed data.
I'm under the impression that DALL-E itself used licensed data as well.
I find some people are comfortable with that, but others will switch to different concerns - which indicates to me that they're actually more offended by the idea of AI-generated art than the specific implementation details of how it was trained.
I then did works in the style of what I’d researched. I trained myself on works I didn’t own, and then produced my own.
I kind of see the AI training as similar work, just done programmatically vs physically.
Certainly a very interesting topic.
I can’t get my head around how far we’ve come on this in the last 6-12 months. From pretty awful outputs to works winning Photography awards. And prints of a dog called Queso you’d have paid a lot of money to an illustrator for.
That's not to say there aren't other ways you might use the direct image (e.g. collage or sampling in music) but you'll likely be careful with how it's used, how much you tweak it, and with attribution. I think the weird problem we're butting up against is that AFAIK you can't figure out post-facto what the "influence" is from the model output aside from looking at the input (which does commonly use names of artists).
I work on an AI image generator, so I really do think the tech is useful and cool, but I also think it's disingenuous (or more generously misinformed) to compare it to an artist studying great works or taking inspiration from others. These are computers inputting and outputting bits. Another human analog would be memorizing a politician's speech and using chunks of it in your own speech. We'd easily call that plagiarism, but if instead every 3 words were exactly the same? Hard to say... it's both more and less plagiarism.
Just how much do you need to process a sampled work before you need to get permission of the original artist? It seems to be in music that if the copyright holder can prove you sampled them, even if it's unrecognizable, then you're going to be on the hook for some royalties.
I don't think that's true.
My understanding is that any image generated by Stable Diffusion has been influenced by every single parameter of the model - so literally EVERY image in the training data has an impact on the final image.
How much of an impact is the thing that's influenced by the prompt.
One way to think about it: the Stable Diffusion model can be as small as 1.9GB (Web Stable Diffusion). It's trained on 2.3 billion images. That works out as 6.6 bits of data per image in the training set.
> My understanding is that any image generated by Stable Diffusion has been influenced by every single parameter of the model - so literally EVERY image in the training data has an impact on the final image.
That's pretty interesting. Need to dig into the math more (lazy applications dev).
If they truly got an appropriate license agreement for every image in the training set then I have no issues with that.
>I'm under the impression that DALL-E itself used licensed data as well.
DALL-E clearly used images they did not have a license for. Early on it was able to output convincing images of Pikachu and Homer Simpson. OpenAI certainly didn’t get licensing rights for those characters.
What ongoing litigation I'm aware of seeks to close that loophole and make fanart illegal, which would be a first step towards also preventing AI art.
Not sure if there would be trademark issues. But that would be the case regardless of how the image was created.
In practice? No, not really. But Pikachu is a copyrighted character and only those with license to do so are actually legally allowed to reproduce Pikachu in media.
Trademarks can come into play like you said, but even just base copyright allows for the ownership of characters such as Pikachu or Batman or whatever.
There is an irony, however, that many of the AI art haters tend to draw fanart of IP they don't own. And if Fair Use protections are weakened, their livelihood would be hurt far more than those of AI artists.
The Copilot case/lawsuit IMO is stronger because the associated code output is a) provably verbatim and b) often has explicit licensing and therefore intent on its usage.
Creating a derivative work of a copyrighted image requires permission from the copyright holder (i.e., a license) which many of these services do not have. So the real question is whether AI-generated "art" counts as a derivative work of the inputs, and we just don't know yet.
>b) often has explicit licensing and therefore intent on its usage
It doesn't matter. In the absence of a license, the default is "you can't use this." It's not "do whatever you want with it." Licenses grant (limited) permission to use; without one you have no permission (except fair use, etc. which are very specifically defined.)
Inspired/trained… I think these could be seen as the same.
If the person directly copied another work, that's a derivative work and requires a license. But if a person learned an abstract concept by studying art and later created art, it's not derivative.
Computers can't learn abstract concepts. What they can do is break down existing images and then numerically combine them to produce something else. The inputs are directly used in the outputs. It's literally derivative, whether or not the courts decide it's legally so.
Goalposts can be moved on whether it has "truly learned" the abstract concept, but at the very least neural networks have the ability to work with concepts to the extent that you can ask to make an image more "chaotic", "mysterious", "peaceful", "stylized", etc. and get meaningfully different results.
When a model like Stable Diffusion has 4.1GB of weights and was trained on 5 billion images, the primary impact of one particular training image may be very slightly adjusting what the model associates with "dramatic".
> If the person directly copied another work, that's a derivative work and requires a license
Not if it falls under Fair Use. Here's a fairly extreme example for just how much you can get away with while still (eventually) being ruled Fair Use: https://www.artnews.com/art-in-america/features/landmark-cop... - though I wouldn't recommend copying as much as Richard Prince did.
> The inputs are directly used in the outputs
Not "directly" - during generation, normal prompt to image models don't have access to existing images and cannot search the Internet.
I would say that abstract concepts is the only thing that computers can learn at the moment, at least until they are successfully embodied.
> It's literally derivative, whether or not the courts decide it's legally so.
To be a derivative work you should be able to at least identify the work it is a derivative of. While SD and friends can indeed generate obviously copyright infringing works (then again so can photoshop or a camera or even a paintbrush), for the vast majority of the output you can at best point out to the general direction of an author or a style.
That's why "fair use" is the key concept here. Under US copyright law "fair use" does not require a license. The argument is that AI generated imagery qualifies as "fair use" - that's what's about to be tested in the courts.
https://arstechnica.com/tech-policy/2023/04/stable-diffusion... is the best explanation I've seen of the legal situation as it stands.
It does not (in US law) if it falls within Fair Use, which is an exception to what would otherwise be the exclusive rights of copyright holders.
I think what you mean to say is that the argument is that both the models themselves and (in many cases) the output from the models, to the extent it might otherwise be a derivative work of one or more the input images, are transformative uses. [0]
[0] https://www.nolo.com/legal-encyclopedia/fair-use-what-transf...
But the term "derivative work" has specific legal meaning. For example wikipedia states: "[derivative work] includes major copyrightable elements of an original, previously created first work (the underlying work)".
I would say that while that might be the case for some AI creations, it is certainly not true for all (or even a majority) of them.
Its kinda like using ffmpeg for vapoursynth for video editing instead of a video editing GUI.
That being said the training parameter/data tuning is definitely an art, as is the prompting.
I turned my dog into a robot awhile back using the img2img feature of Stable Diffusion and the results were pretty amazing![1]
Say you generate a picture with midjourney - who is/are the closest artist(s) you can find for that picture?
Not the AI, not the prompter, so the closest artists you can find for that picture are the ones who made the pictures in the training set. So generating a picture is outright copyright infringement. Nothing to do with unfairness in the sense of "artists get out compete". Artists dont get out compete - they are stolen.
Why is it you discount the creative input of the user? Are they not doing work by guiding the agent? Don’t their choices of prompt, input image, and the refinement of subsequent generated images represent a creative process?