this post was submitted on 02 Aug 2025
39 points (91.5% liked)
MeganeMoe
521 readers
16 users here now
Megane is the japanese word for eye-wear, glasses. A character wearing glasses is often meant to communicate intelligence, maturity and dignity. Glasses are considered by many to increase the appeal of a character, and some series use sets of glasses as a major plot device. Persona is perhaps best known for this.
- Posts must feature someone wearing glasses.
- No nudity. Lewd art is fine, but mark it NSFW.
- If posting a more suggestive piece, or one with simply a lot of skin, consider still marking it NSFW.
- Include a link to the artist in post body, if you can.
- AI Generated content is not allowed.
- Positivity only. No shitting on the art, the artists, or the fans of the art/artist.
- Finally, all rules of the parent instance still apply, of course.
SauceNao can be used to effectively reverse search the creator of a piece, if you do not know it.
You may also leave the post body blank or mention @saucechan@ani.social, in which case the bot will attempt to find and provide the source in a comment.
Find other anime communities which may interest you: Here
Other "moe" communities:
- !fitmoe@lemmy.world
- !murdermoe@lemmy.world
- !fangmoe@ani.social
- !cybermoe@ani.social
- !streetmoe@ani.social
- !midriffmoe@ani.social
- !kemonomoe@ani.social
- !thiccmoe@ani.social
- !officemoe@ani.social
- !gothmoe@ani.social
- !militarymoe@ani.social
- !morphmoe@ani.social
- !smolmoe@ani.social
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
That's AI with some tracing over it, if you look at the folds in the shirt, there is odd discoloration
It's apparently the other way around.
I gave Redpostit a once-over a while back. They did not exhibit the usual signs of "prompt-artists". Because they aren't one.
I didn't realise they do use AI tools though. It seems Redpostit draws lineart and coloring, then shades using AI. To me it's fairly obvius the image is further adjusted after that as well.
I'm unable to find a first-hand account of their workflow. I'm not sure what kind of model is being used.
I'll have to think about whether I will remove their work and disallow it in the future. But my first instinct is to compare this to what Joel Havier does for his animations.
Redpostit does not have a rate of output typical to AI "prompt-artists" and most of their work is consistent in terms of lineart and charachter style.
I am willing to believe they do in fact put in the majority of work in terms of creative labour.
I don't know man, this sure looks like an AI generated girl with some coloring and textures on top
It looks like a trace over a movie still where they drew them in a new outfit. This character is not AI. Chel from Road to El Dorado.
The character is from El Dorado yes, you can generate AI images of existing characters. It's not a traceover of a movie still
I don't use "vibes" to tell.
Come back with something that proves it to be generative, or just voice your opinion that the no-AI rule should apply to assisted works as well.
It sounds like you're also working off of your own assumptions though. What constitutes "proof" to you? Also if the rule is no AI... Then this applies, it uses AI.
The rule is no generative AI.
My one assumption, currently, is that I believe what the artist has said about how they produce their content.
I have not made my mind up about whether artists like Redpostit, who use AI tools in their workflow, should be allowed. It has not come up until now.
As of now I am inclined towards allowing it. But I want to investigate Redpostit more thoroughly, as well as find some other artists that do something similar, to greater or lesser extents. As well as look up the AI models that'd be used to perform only certain steps of the work, such as shading.
If you'd like to sway me, please do so without being adverserial, without assuming someone is lying without proof of that, and after you've familiarized yourself with the current rules.
And please drop your apparent assumption that I've already decided.
My very first reply explicitly states I'll have to think about it.
My personal reason for disliking AI is the tremendous environmental cost, it destroys freshwater and it is for the most part fueled by mega data centers fueled by carbon emitting fuel sources.
Additionally all widely available AI libraries have been largely trained from stolen work from artists without their permission. Meaning even if this person is using AI "just for the shading" it is still piggybacking off the stolen work of other creators, and, depending on if they run this AI generation locally or through some service, also contributing to the tainting of fresh water and pollution via carbon emissions.
In my opinion I also see this as generative AI, these line drawings are fed through the same ai generator as fully generated AI content, and it is generating a large aspect of the art.
Overall I just don't think this is something that should be promoted because it's ethically dubious at best. Especially since the creator seemingly is not very open about the use of AI.
The most likely tools being used are Krita AI, or copainter.
Krita AI is a frankly impressive plugin for Krita that allows for a "collaborative" drawing process, where you are able to manually iterate on every conceivable detail, until essentially no AI brush strokes remain. And it runs locally.
But it can also go from an empty canvas to a complete piece based on nothing but a prompt. Or anything in-between. How much the AI does is entirely up to who is using it. It actually looks like a potentially amazing way to learn, if you're willing to turn down how much help you're getting over time, and not take credit for something you didn't produce. To develop your own illustrating ability and eventually take over entirely.
Copainter takes lineart, and produces what looks like a shaded 3d model. There are potentially ethical ways to train such a model. If the dataset was created by essentially turning a bunch 3d models into line-art, easily done in various automated ways, you might use the resulting dataset to train a model to reverse the process even with novel images.
If this was how it was done, I still find it unlikely the 3D models used for it weren't just scraped, but 3D assets are a bit more difficult to find than 2d assets.
The more active user-base seems to be around Krita AI. The artists I could find using it seem to be engaged in a wilful self-delusion that using their own art to "train" the model themselves, circumvents the IP theft aspect of AI training. But they're not really modifying the model. They're just fine tuning it using LoRA and pretending it's more than it is.
The results are still enabled by the large scale theft of uncredited work for training the base model (Stable Diffusion).
So unfortunately, I can't compare these tools to radiance fields, SwitchLight or EbSynth animation.