Based on the general consensus users are leaning towards ensuring that AI posts are marked as such and to limit the amount of them. Adjustments to the sidebar rules will be made stating that AI posts must be labeled in the title as Lemmy does not currently offer a way to provide flair. If we see a significant amount of AI posts then we will reassess banning them from this community.
Yeah restrict AI posts. I think cuteness needs to have some notion of soul behind it to be cute and AI images are just creepy and soulless.
I don't want to gatekeep what is and isn't aww, but I definitely don't appreciate the realization that something fake was slipped in right under my nose without warning. AI may be harder to get good results from than I imagine, but if it's not being presented as exactly what it is upfront, it's naturally going to fool a lot of people into thinking they're seeing an extraordinary image.
Come to think of it, this doesn't just extend to AI. If someone was trying to pass a photo realistic render or heavily Photoshopped pic off as real (and I mean on the level of replacing, grafting and moving parts of multiple images to make something that gets attention for how surprising or special it seems) then I would feel the same way about that. It's about honesty. I want to see all kinds of mediums on here allowed to shine, but not if it's done through obfuscation of the medium used.
And also, AI stuff just skeeves me out. I want to be able to filter it out, or just know that there's a rule against it and another community for that content. There's something deeply human and heartwarming about an "aww" to me, and if it's not real, or at least presents itself honestly in a way I can appreciate the love that went into the craft, it's no longer aww.
Perfectly said.
I understand and struggle with the idea of possible dishonesty. TBH I only saw this post in my local feed and commented about why it doesn't make sense to ban this. The negative culture against this new tech is more dangerous than even the tech itself. This is why I posted and have kept watching this thread. The images generated are based on real images. It is challenging for a beginner to learn how to prompt for both the composition and photorealism. Composition is mostly done with positive prompting; saying what you want to see with the right collection of words and phrases. Photorealism requires both positive prompts about lighting, focus, and image characteristics, but also a negative prompt that removes large portions of the dataset, like CGI, cartoons, anime, etc.
I don't know of a single reason I would ever waste time on SD trying to generate images to post here. The reason I am trying to say it matters is because everyone needs to be aware of what is and is not possible in this space and stay up to date on it. The tech is moving extremely fast. The there are a lot of very subtle issues that creep into Stable Diffusion outputs. The more detailed and specific the prompt gets, the harder it is to prevent these small errors. After I played with the technology for a few days, I quickly learned to spot many of the errors. Learning this skill right now, to be able to spot the subtle AI errors in images may be really important in the near future. After learning this, I can spot a lot more images that are text2img generated and I would not have known otherwise.
This technology is not a fad. It is not going away. It is a very powerful tool. Calling it "AI" is rather dumb. It is not AGI (general) in any remote sense. This is probably the most disruptive tech since the internet of the mid 90's or capacitive touch tech enabling effective smart phone interfaces, or crypto currency making billionaires.
It doesn't matter where you get familiar with text2img nuances, you need to see the almost good, good, virtually indistinguishable, and truly indistinguishable potential to know what is really possible here. You really need to see the level of complexity that is currently possible in a prompt. There is a limit here. If you know these types if limits, you will have a much better sense of truth in this space. Everything visual should be met with skepticism now. Relegation of this tech to the engineers gives them extreme power over you now and even more in the near future. Please take the time to learn about this and be familiar with it.
As someone that has messed with Stable Diffusion for generating images, this is a non issue. AI images are challenging to prompt well and generate results. Good results are the result of clever prompts and what really amounts to a new form of artistry. The hype over AI is mostly due to mainstream garbage hype media and people that can not distinguish between technology and magic. If anyone opens Stable Diffusion and tries to just write a prompt, the output is usually garbage. If they are super lucky, the first result may work, but it will go downhill from there. Good prompts take practice, research, and lots of extra software tools. It is a skill all its own. If you are going to restrict AI for anything, it should be limiting it to a tag in the title, must include the entire prompt and seed, and the positive prompt must match the image reasonably well. This would limit post submissions to real artistry and it would enable anyone that is interested in this technology to reproduce the results and explore them more.
must include the entire prompt and seed
I like this one, better than straight up restricting it and maybe, just maybe, we'll get quality aww's. If we don't, then we can ban it.
I think this would be the best compromise. Banning stuff has historically not worked well, so embrace and enforce should be the chosen avenue.
In my eyes, the question is more if this is even an issue. I mean, if this community gets flooded by AI stuff, it can be forbidden at any time, but as long as there is only the occasional AI post (that makes you go aww), why artificially (hah!) Limit ourselves.
I think this is a healthy approach. Allow it (temporarily), make sure to maybe mark ai stuff and let's see...
Yes, please restrict
I think it should be restricted. However, if it's decided to be allowed, it should be tagged as AI.
I remember the problem on reddit was people posting needlefelt dolls as real animals and that would piss everyone off. AI can be Aww, but it should be tagged as AI and be removed if not tagged.
I’m not a fan of Ai generated images as straight content, I’m a little more fine with the ChatGPT at work but even then that appears to be a bit of a gray area when it comes to the training data that any AI uses.
I’m gonna have to pass on AI content for the simple worry of the wrong type of content saturation.
You can either go full bonzai and aggressively trim any little branch that pokes out of place to try to keep a meticulously maintained tree at all times, or you can just let the tree grows as it will, and if a branch becomes an obvious issue, then just cut the entire branch and graft it somewhere else.
If I was a mod here, I would do the latter, maybe even setup an /c/AIAww or whatever in anticipation for what might come.
That's probably my laziness speaking.
Is there a way to have them marked as AI like NSFW, or some sort of hashtag? That Way people can easily filter them out if they don't want to see them.
If it's so good that it looks like a real photo I suppose it's a moot point, but if it looks less than real then I would probably downvote that.
Please don't allow AI posts. If we start having AI posts this place will basically turn into a fan art page.
No AI please - it just dilutes and lowers the overall quality of the community. I'd much rather a handful of genuine posts/comments from real people than a flood of hallucinations that range from being utter gibberish to making you question your own sanity.
Maybe comoderate something like c/AiWWW to have a clear 0eperation but if people want they can get over the info a community that is specifically for it
The problem with that is we're already such a small community, something so niche as cute ai generated images only would be so microscopic that it may as well not exist. Back on Reddit they had the userbase to support such niche interests, but the fediverse isn't there yet.
I feel like AI posts in this context are no more harmful than Photoshop. It's not really a community about artistry or appreciating the authenticity of the post.
I would say let them. If people feel deceived by this, but not when someone claims a drawing or picture is theirs when it's not, that's a them problem.
Compromise? Ai art must be tagged so.
There’s a lot of creative ai art out there, is it as impressive as say, a painting? No. But is it still art? Sure. What is “art” anyhow?
AI content should be limited to AI communities. Period.
One per week per user for completely generated images.
Filtering an existing photo is okay, but using multiple filters to post multiple images is not. Only one post per distinct source image.
From how I see it, they're fine as long as they're labelled as AI posts.
It's fine as a concept, but I don't really want my feed of actual aww-worthy images diluted with ai generated pics.
Echoing the sentiment of others, it should be restricted and/or tagged as AI. Preferably it should be limited to AI-specific communities.
I'm cool with AI. I've been having a blast playing with stable diffusion
aww
A place with minimal rules for stuff that makes you go awww! Feel free to post pics, gifs, or videos of cats, dogs, babies, or anything cute and remember to be kind to others.
AI posts must be labeled [AI] in the title and are limited to one per week.
While posting and commenting in this community, you must abide by instance-wide rules: https://mastodon.world/about
- No racism or bigotry.
- Be civil: disagreements happen, but thatdoes not provide the right to personally insult others.
- No SPAM posting.
- No trolling of others.