Should AI-generated 'stories' be banned? Just askin'!!
Should AI-generated 'stories' be banned? Just askin'!!
In my opinion, categorically banning "AI generated" stories, no.
If someone deliberately uses AI to copy someone else's work, that can and should be banned, but you wouldn't be banning it because they used an AI.
If the story is unreadable, maybe, but again you aren't banning the story because AI was used, but because of the low quality result.
Stories? No. I like that they got their own category, so I can simply avoid them.
I think users who refuse to mark their stories as AI-generated should get banned. And, yes, that includes the ones who "only" use AI to write the filler parts between the sex scenes. You know who you are.
I disagree with that entirely, AI-generated and using it for padding are two different things. An AI creating the entire thing, then yes, that needs to be labelled. An AI to help pad out a story, is no more than an editor. I can normally tell which one's are AI generated, even if they aren't labelled, I just avoid them. I don't think people should be banned for it.
Wow! Yeah, sorry, but I disagree. Wholeheartedly. Editors are something COMPLETELY different from AI, because they understand what they're doing. And if you have to rely on AI to "pad out" your story, maybe you're just not the writer you think you are.
I just had that discussion with someone on Reddit. They thought the AI-generated stuff sounded better because it was "verbose" and "grand". Problem was, when I actually looked at it, the AI had "padded" the story with countless metaphors that were completely out of place, didn't lead anywhere, and added absolutely nothing to the story other than SOUNDING sophisticated. It made the whole thing exhausting to read!
The big problem with AI is that it doesn't understand what it is writing. It just lines up words it has been trained to line up. But it doesn't get WHY it lines up THOSE words. There's no MEANING behind it, whatsoever.
So, you have bad writers who can't maintain a narrative, and lazy writers who can't be bothered to set their scenes, rely on AI to do it for them, resulting in absolute crap stories they flood the site with, pulling down the overall quality of the site and therefore reducing our audience.
There is no shame in being inexperienced. There is no shame in not being perfect. But if you use AI to write your story for you instead of trying to improve and grow, then, sorry, but I feel like you should be kicked out. Go and sell Ghibli-Style AI "Art" on Twitter.
But therein dear reader, lies the issue between knowing the different between padding and dense.
But therein dear reader, lies the issue between knowing the different between padding and dense.
...what?
Is that why you use AI? Because THAT'S what it looks like when you try to form sentences yourself?
Pulling down the quality of the site?
Lets be honest here. Some of the work on erotica sites is great and has been since these sites started, and some of it has always been extremely badly written.
If anything, AI tools would up the quality for a number of stories where the 'author' has a neat idea but just doesn't know how to execute on it.
Half the stories written are personal novel projects, works of great emotional interest, and so on.
And the other half have always been basic stroke material written and meant to be read 'one-handed'.
And there's work in both categories that could use help, but in the second category nobody's either writing nor reading to experience literary masterpieces.
Let those folks have their fun I say.
I don't and can't write stroke material personally. If I used an AI I'd probably use it for the sex scenes, which is what often stalls my projects. But they tend to filter against that and I personally just don't want something else's style inside my work.
But when I'm looking at a stroke story that's not what I care about. I have an itch, it's scratching it.
I could care less if it was written by chat-bots.
If we care about banning improper content - start with anything that's fanfic or otherwise uses someone else's intellectual property. That's a lot more improper than stuff a chatbot spits out. But nobody's ever cared to clamp down on that stuff.
If anything, AI tools would up the quality for a number of stories where the 'author' has a neat idea but just doesn't know how to execute on it.
That discussion has been had on... uh... another big site for erotic literature, about half a year ago. And, back then, I was completely with you. I saw the barely readable masturbation fantasies submitted by the dozens on a daily basis, and thought AI could only improve on them. But when the AI-work then actually started flooding the site, I was proven wrong.
All AI does is make the stories readable, but they don't magically become good, while giving the "authors" false confidence. Most of these stories are exhausting to read, but the "authors" think they're great, and flood the site with even more of their crap now.
That's my problem with AI. As long as "authors" mark their Ai-generated, low-effort, pseudo-sophisticated drivel, I can avoid it. But once they stop doing that, it'll become harder and harder to find actually well-written works, and actual storylines written by a person that can maintain a narrative, and actual character developments written by people who don't forget details after three chat messages.
An AI to help pad out a story, is no more than an editor.
If the AI was being used to check spelling and grammar, that would be true. (Is current AI even able to spot things like continuity errors and misspellings of character names?).
But writing padding is adding new content - that's not strictly editing. "AI, write me a description of a posh restaurant in Paris" is new content that the author should have researched and added themselves.
AJ
AI programs are designed to accept 2 things as input. First, a request and description of the story to be generated. Second, a database of existing stories. Then it patterns the new story based on its database of existing stories.
If the existing stories were rated as poor, then the new story is very likely to be rated as poor.
Do we really need to be overwhelmed by a flood of poorly written stories?
Over the past few weeks, I started reading several stories. In two of those stories there was a very noticeable error in the first sentence of the story. The rest of the story was riddled with other errors, such a referring to a male (or female) character as she (he).
The problem with trying to ban Ai-generated stories is how do you know that the author used AI to generate the story if they don't admit to using AI.
Stores about AI are fine, as those have always been both popular and powerful. Stories which were edited with the assistance of AI are, problematic, yet the intent is good, and their stories are probably better for it, but once you're stories are submitted, it also means the AI systems can then access them, meaning you very well might encounter future copyright violations (either yourself or others who's work includes direct sentence fragments, sentences or even entire paragraphs of yours.
Alas, once we free the genie, we're all subject to it's many whims and quirks, so best leave those genies and demons be.;)
Should AI-generated 'stories' be banned? Just askin'!!
Some of those posted on SOL are pretty decent IMO (but some aren't). Presumably that's a reflection on both the quality of the AI and the ability of the 'author' who operated it.
I don't think there's a strong case for banning AI generated stories at the moment. That might even be impractical because it's impossible to know how much the AI has generated - whether it's fillers, wallpaper, specific scenes, or the whole story.
However if the quantity of AI-generated stories becomes a flood, making it difficult for readers to find stories with original content, that should prompt reconsideration.
Who knows, using AI to generate stories might achieve mainstream acceptance, like using spellcheckers.
AJ
Some of those posted on SOL are pretty decent IMO (but some aren't). Presumably that's a reflection on both the quality of the AI and the ability of the 'author' who operated it.
Assuming Lazeez correctly flagged it.
I tried one story with the AI tag and sent the author a note asking him about his AI experience, and how he was able to generate sex scenes with AI. He never answered. So I have no idea if AI actually generated that story.
There are AI engines that will generate uncensored NSFW content. Of course they generally aren't free.
One I've been playing with is https://novelai.net/ It's completely uncensored but you can't do much with a free account.
Assuming Lazeez correctly flagged it.
Like most SOL categories, those are writer chosen, not system assigned, so you'd assume the author knows how it was developed.
That said, given how prevalent AI is, and how horrendous the modern AI tools are, I'd likely never read anything listed as "AI" generated, and given the majority of the current SOL stories, I'm barely following many stories as it is. But then, the few that I am, do not qualify for FineStories either.
Like most SOL categories, those are writer chosen, not system assigned,
Lazeez said he tagged most of them.
I stand corrected then. I guess I'd heard that, yet I'm more used to the older systems than the more recent changes, since I haven't posted much lately.
I have one that is a combination of AI and my writing, because the AI couldn't do enough or lost track so I had to fix. It's not the worst thing any of my pen names have written, but nowhere near the best
I'd never criticize that approach, what I do bitch about is when someone just types in an AI prompt and simply assumes it's adequate, as often it's not. You're definitely using AI correctly, but again, "AI" is an overly generic term, so you're never sure what it's referring to, so it's hard to determine what's fitting and what isn't.
But again, I keep on ranting, rather than just shutting up and dropping the topic.
No worries. The story published under this name was to test out an AI erotica app. It was an okay experience, but also as someone who has many stories (not novels though) on here, and who sometimes finds adhd and other things interfering with writing, I thought I'd see if it could jump start something. I'll be intrigued more to play with it in the future, but it's been mostly a disappointment so far, though I've also only been using free things
Assuming Lazeez correctly flagged it.
I've noticed a couple of stories recently without the AI tag. Both look as though the authors gave AI the plot of a chapter then let it loose, the results being excruciatingly florid (not the authors' normal styles) and repetitious, repetitious and very repetitious.
One of the stories even has a reasonable rating, but I suspect that's due to fanbois of the author having a previous stellar-rated story.
AJ
A lot of people on here cannot write, a lot of people on here don't have time to edit. Some people are disbaled and literally cannot write. AI will help them, as it helps me. I think as long as they're not just ripping someone off, or being lazy, then its fine. No more so than sudowrite, grammarly or any of that.
Amazon KDP, during publishing, now asks if you used AI. And when you say yes, it asks you to choose how much (how extensive). I think it said it's gathering information, but I'm not sure of the reason.
When I said yes, there were three uses of AI: 1. the text, 2. the image or illustrations, 3. (I forget what it is).
I said yes because I generated my cover image with AI. Then under #2, I chose the least use (something like few changes or modifications).
The problem I encountered was that when I clicked to go to the next page it stopped me, telling me I had to correct errors on that page. The errors were that I didn't choose a "how much" answer to the AI #1 and #3 questions.
Duh, no text (#1) was created with AI. And I didn't use AI for whatever #3 was. So I was stuck. I had to go back and tell Amazon that I did not use AI. How stupid is that of them?
Sounds typical for them. I had a software download that defective and they recalled it. They credited my card and told me to just delete it off my computer, then they emailed me a Prepaid UPS label to return it. That was fine since it didn't work at all.
Wildly off topic, but ...
Someone I know, who works part-time at an Amazon return counter, bought a glass water bottle from them. It broke (without any mishandling on her part) almost immediately. Clearly it was a manufacturing defect.
Amazon told her to return it. From working for them, she knew Amazon does not take returns with any broken glass in them.
They sent her the refund immediately, after she reminded them of that.
They might (maybe, possibly) be wanting to flag things so nothing written using AI gets used to train AIs. That's going to be the next big struggle - keeping AI outputs out of AI inputs. We're already losing that struggle, and it's screwing things up.
They might (maybe, possibly) be wanting to flag things so nothing written using AI gets used to train AIs.
The point was, they shot themselves in the foot whatever their motive was. I was being honest and answered yes to the image part. But by doing so, I had to say I also used AI for the text (and whatever #3 was) which I didn't. So I ended up answering no for any AI.
I mentioned it in this thread because it's a publisher asking questions about the author's use of AI.
My comment was in response to 'I'm not sure of the reason', not in response to the poorly designed questionnaire.
I visit a couple of discussion forums. Fairly often, someone suggests banning AI posts.
For some reason, the moderators/owners of the forum are adamantly against banning AI, and won't even allow discussion on that subject.
All that does is make me wonder how many of the posts may be generated by AI to make the forum look more popular.
Disclaimer: I didn't bother reading the following article, but thought it fits into this thread. The article title is:
"College Students Are Sprinkling Typos Into Their AI Papers on Purpose."
and the first paragraph is:
"To bypass artificial intelligence writing detection, college students are reportedly adding typos into their chatbot-generated papers."
https://www.yahoo.com/news/college-students-sprinkling-typos-ai-180132859.html