AI Allegations Thread

[alohadave]

She has clearly taken a harsher stance and it sucks for some people so I truly sympathize with them. But there is also one hard truth here. Just because some of those people claim they are not using AI doesn't mean they are not using it. It feels weird to defend Laurel's actions for once, but I can see how hard her choice is. She can either take a hard stance and take down all suspicious stories and in the process take down some legit stories too, or she can be cautious and risk the website being flooded with AI content. It really gets down to what is important to her personally as a website owner. She chooses the policies and that is fine with me.

Sure some people are using AI.

I'm not disputing that she has a hard choice. But there will come a time when people have had enough and leave because they are tired of being jerked around. We are already seeing it here.

When you do what she says, the way she says she wants it and get kicked back multiple times, you feel like it doesn't matter what you say, so why bother submitting here?

 

[AwkwardlySet]

How else should I take it? You use the superhero argument from The Incredibles and wonder that I inferred that from your word.

I had to google The Incredibles. I have no idea where and how that line is used in this animated movie, but I can tell you that they didn't come up with it. That thought has roots in philosophy unless I am severely mistaken. And exclusivity is not the point of that line, at least not to me. It's about the value of an artistic creation. Either way, I am in no mood to respond to belligerent posts.

 

[AwkwardlySet]

Sure some people are using AI.

I'm not disputing that she has a hard choice. But there will come a time when people have had enough and leave because they are tired of being jerked around. We are already seeing it here.

When you do what she says, the way she says she wants it and get kicked back multiple times, you feel like it doesn't matter what you say, so why bother submitting here?

It is like that already for some of us, on some other issues. I'll just say what I said once before. Lit won't get better for authors until there is some healthy competition out there. Right now, that is not the case, so yeah, we can bark at the moon all we want

 

[alohadave]

I had to google The Incredibles. I have no idea where and how that line is used in this animated movie, but I can tell you that they didn't come up with it. That thought has roots in philosophy unless I am severely mistaken. And exclusivity is not the point of that line, at least not to me. It's about the value of an artistic creation. Either way, I am in no mood to respond to belligerent posts.

I'm not being belligerent. Take it however you wish. But I'm asking you a question in good faith.

Does your self worth rely on there being a limited pool of authors? Or do you feel value by creating for the sake of creating?

Does it matter to you if a million other people do the same thing as you? Or by being part of a small club of people who do this make you feel more special about your writing?

 

[AwkwardlySet]

I'm not being belligerent. Take it however you wish. But I'm asking you a question in good faith.

Does your self worth rely on there being a limited pool of authors? Or do you feel value by creating for the sake of creating?

Does it matter to you if a million other people do the same thing as you? Or by being part of a small club of people who do this make you feel more special about your writing?

It is not about my own self-worth, it is about the value of the creation.
I don't like this comparison, but we can talk in the sense of market value even. Something you can get anywhere and easily - let's say simple rocks, has no value. Something that is rare or hard to obtain, or both, has value. I don't want to spend time creating rocks, it is as simple as that.

 

[alohadave]

It is not about my own self-worth, it is about the value of the creation.
I don't like this comparison, but we can talk in the sense of market value even. Something you can get anywhere and easily - let's say simple rocks, has no value. Something that is rare or hard to obtain, or both, has value. I don't want to spend time creating rocks, it is as simple as that.

I say that the act of creation is what gives it value. It may not be marketable or sellable, but that is different than it's inherent value as a creative work.

My creative background is photography, so I'm very familiar with this kind of viewpoint. When I travel, I go to tourist spots and take the same pictures of the pyramids, or the Eiffel Tower, or whatever. They might be very similar and taken from the same vantage point, but they have value to me. I took the picture and it is mine. Does anyone else care about that picture? No, not really. But that's not the point. The point is the act, and showing that you were there.

Realistically, none of my pictures (and by extension, my stories) are anything special or cared about by more than myself, family, and some friends. But I make them for me. If other people like them, that is a bonus. But not being unique doesn't make me value what I've done any less.

This is from Full Metal Jacket and I use this quote all the time because it fits so many things: "There are many like it, but this one is mine." And that's all I need from my work.

 

[theWollstonecraftWoman]

And the battle continues.

 

[TheArsonist]

No, the latter doesn't necessarily follow from the former. The other possibility is that the rewrite constitutes a derivative work, in which both the original author and the rewriter have copyright interest, and neither would be able to post it to Literotica without the other's permission.

True enough in theory, but courts have held that machines cannot hold copyright, the companies behind these rewriting tools presumably disavow any copyright claims on their part, and I think it is absurd on its face to suggest that any of the authors of the pre-training material can make a copyright claim based on having used e.g. the phrase "like a weary pilgrim" before—the nature of the task means that there is no risk of stealing substantial, copyrightable parts of any particular work. So that just leaves the author of the text that was revised.

Not that an excerpt that short would be likely to constitute copyvio, but hopefully that illustrates that GPT can indeed learn the source material so closely as to be effectively "making a copy". I expect most of the stories it's trained upon would be less perfectly memorised, but it doesn't need to be perfect to be copyright violation.

"Essentially" and "fancy version" are doing a lot of heavy lifting here. 8letters' statistical analyses don't have the capability to reproduce entire chunks of the stories they're based on.

Which, as seen above, they can do.

I absolutely agree that LLMs can commit copyright violations. My point is that just having the capacity is not a violation in itself; the copyvio doesn't take place until they actually output something that violates the copyright.

With a story generated from scratch, that is a non-zero risk (especially if the prompts reference some particular author or story), and so I see a good argument for disallowing it. But in my view that risk is not present for original stories that have merely been rewritten with AI support.

 

[theWollstonecraftWoman]

Your belief won't sway Laurel's views.

True enough in theory, but courts have held that machines cannot hold copyright, the companies behind these rewriting tools presumably disavow any copyright claims on their part, and I think it is absurd on its face to suggest that any of the authors of the pre-training material can make a copyright claim based on having used e.g. the phrase "like a weary pilgrim" before—the nature of the task means that there is no risk of stealing substantial, copyrightable parts of any particular work. So that just leaves the author of the text to be revised.




I absolutely agree that LLMs can commit copyright violations. My point is that just having the capacity is not a violation in itself; the copyvio doesn't take place until they actually output something that violates the copyright.

With a story generated from scratch, that is a non-zero risk (especially if the prompts reference some particular author or story), and so I see a good argument for disallowing it. But in my view that risk is not present for original stories that have merely been rewritten with AI support.

 

[theWollstonecraftWoman]

@TheArsonist I would also add that Laurel or her husband monitor postings here and by posting your use of AI to rework material you may draw a big BULLSEYE on your work.

 

[TheArsonist]

Your belief won't sway Laurel's views.

No, but I'm convinced that the policy is completely unworkable in practice, and she'll either have to change it or the site will be shunned by most new authors.

 

[Throwaway256897q6r72]

@TheArsonist I would also add that Laurel or her husband monitor postings here and by posting your use of AI to rework material you may draw a big BULLSEYE on your work.

Pretty sure that's what happened to me today. I mentioned that I've used online sources to help with grammar and spelling and the next time I came onto the site, most of my stories had been pulled

 

[nice90sguy]

Pretty sure that's what happened to me today. I mentioned that I've used online sources to help with grammar and spelling and the next time I came onto the site, most of my stories had been pulled

I don't think it's "mentioning" that got you rejected, more likely the gramatical "fingerprint" that your automated grammar helpers introduced. Stop using them, maybe?

 

[TheArsonist]

@TheArsonist I would also add that Laurel or her husband monitor postings here and by posting your use of AI to rework material you may draw a big BULLSEYE on your work.

Just because I think it's foolish to ban it doesn't mean I do it. (I also believe in legalizing marijuana even though I don't do pot.) For one thing, a big part of why I think it's a bad rule is that people are being falsely accused of having used AI tools.

I do not myself use AI to rework material. For one thing, nothing I'm writing for LE is at the stage where it would be relevant, but as previously mentioned, I don't think it improved Dybbuk's example, either.

I'd certainly be open to run an AI-powered check to flag any grammar, readability or stylistic issues (such as overusing or repeating certain words or phrases—like the way I used "for one thing" in both the paragraphs above), but I'm precious enough about my writing that I don't want somebody or something else to fiddle with it unless I've made an outright mistake. I gave up several writing gigs because I couldn't abide editors making changes without consulting me.

 

[Throwaway256897q6r72]

I don't think it's "mentioning" that got you rejected, more likely the gramatical "fingerprint" that your automated grammar helpers introduced. Stop using them, maybe?

So, I either get rejected or have poor reviews for making a few grammar mistakes that I don't catch or I get rejected for using a helper? Doesn't make much sense to me.

 

[nice90sguy]

It might also help to change your name to something that doesn't look like compter-generated strong password.

(Waits for humerectomized response)

 

[Throwaway256897q6r72]

Right! Haha I honestly wasn't expecting to keep writing stories after I publish my first few elsewhere and when I started to post them here, I just kept the name as to not confuse myself.

 

[ElectricBlue]

I can't see how playing with my original paragraph and making some adjustments even comes close to stealing. Perhaps you have an image of someone asking the LLM to write a story for them, but that's not what I do, not even close. In any case, I have discussed this topic enough.

I have no work here, so there's nothing to take elsewhere.

You seem to be missing the point. The AI tools you are using to "enhance" your fictional writing have been, more than likely, trained on copyright content, illegally obtained. Using the tool, therefore, is building on a body of quicksand, ethically, morally, artistically.

The best way to avoid the issue is not to use the damn tools at all, and write your own words yourself. From the samples you've given, I think you're fooling yourself that they're "better". They might be "different", but define "better".

Trust your own writing, not some digital distillation. Write and edit yourself - that's when your writing will shine. Meanwhile, you're merely polishing a pastiche, and to be honest, it shows in the examples given. They're no-one's words, they're everyone's, but they're not yours.

Trust yourself. Why do you need a crutch?

 

[mirafrida]

This is the revised version with artificial assistance:


And this is the original paragraph:


The changes are mild but undoubtedly worthwhile.

I gotta say, for my money the original was way better than the "ai enhanced" version

 

[theWollstonecraftWoman]

Okay, calling out site owners isn't the way police is changed here. Not only that, but they own the site, not the readers, not the writers, and our opinion of their policies isn't something they need worry about. There are dozens of sites you can post stories on. Some with few rules about much of anything.

I'm dropping out the conversation as it is unproductive and those hell bent on lambasting the owners do not represent my views. I'm certain my work would read better if I let AI rework it. But won't do that, because that isn't how a person learns by having software change their words.

 

[ElectricBlue]

I mean you make good points here but some people are saying they didn't use anything and they're getting all their stuff removed while others readily admit they use writing tools and sometimes take the suggestions, and their work is still up

I'm responding only to the poster who has been saying, "I use it because I think it improves my writing, why can't I use it for that?" I'm asking, "Does it improve your writing?" and I'm answering , "No, it doesn't."

Your comment is valid, but it's another issue - and I don't know the solution for it. But I think I recognise AI when I see it, and I reckon, so does Laurel.

If existing stories are being removed, it's because they're being reported. I don't agree with that, reporting a story because of its style.. Reporting borderline content, yes, but not for bland prose.

 

[NoTalentHack]

ok you say you use a writing aid and take its suggestions sometimes but that's exactly what the new guidelines say is against the rules. I'm not saying what you're doing is wrong or right but have any of your stories been taken down since you posted this? Other people are saying they didn't use anything but spelling check and they're getting everything pulled. I know it's the owners site but maybe it would help if the guidelines say that new writers are subject to heightened scrutiny over this issue

I started using PWA June 5th. Since then, I've posted 14 stories and 1 essay, none of which encountered any issues in the queue, and two of which came in (probably, it's hard to be certain) fourth or fifth in contests. When I say that I take its suggestions "sometimes..." Hang on, let me give you an example.

This is from a story I'm editing now, in a section I've already gone over multiple times; I edit sometimes when I don't have the energy to write, so the early parts of my stories tend to be a lot "cleaner" than the later bits. Regardless, any additional changes are likely to be minor, and I've taken about all the suggestions from PWA I'm likely to take.

View attachment 2294304

The blue squiggly underline is Google Docs' sole (and incorrect) contribution. The blue underlines are from PWA, and they represent places where it thinks I've screwed up grammar or punctuation; PWA is either incorrect in these places or IDGAF. The yellow lines are places where it thinks I can improve my writing in some way: use of adverbs instead of stronger phrasing, extraneous words ("different possible" is one of the choices it flagged in this case, since they either could be used; that's probably true, but I like it better that way), passive voice, "glue" words or phrases ("even" or "just" as an intensifier, or "for that matter" as a stylistic choice), and so on.

Like I said, I take, in a 10K document maybe 10-15 of these suggestions and swap them out for something else. But the rest? I let them ride. I do use some other tools, too; this is the bare minimum of what PWA can use. But what I don't do is this:

View attachment 2294308


View attachment 2294310

I like my stupid, lazy phrasing. "Very obviously" is more conversational, which is... well, it's my voice. That's how I think, and it's how I write. I'm not saying someone could look at something I wrote and go, "Yup, that's an NTH story," but I think the people that follow me could read one of my stories and say, "Wait, this is an NTH story?" if I did something like that. Especially if I made a big shift to that.

If someone is using PWA and taking every one of those yellow underlines and saying "rephrase this," or even tinkering with them on their own before submitting until there are no yellow underlines? Yeah, they're going to have a voice that's recognizably not theirs.

For the curious, here's a mess of writing near the bottom where I haven't had a chance to edit at all. I have to get it all done by tomorrow! God help me.

View attachment 2294311

There are LOTS of things wrong with this section, but not that many more things that PWA caught. The red underlines are misspellings or words it doesn't recognize. The purple is a passive phrase it thinks it can rewrite; I may or may not change that, because I kind of like the passive voice there.

But there's stuff it hasn't caught, too: repetition of words, description of action that could be more vivid (because it recognizes adverbs as being signs of "weak" verb choice, but it sees lack of detail as clean, rather than boring), screwed up character indicators (there are four characters in the scene, but it doesn't "know" that), and so on. That's the other thing: these tools know what they know, but they don't know what they don't know.

This is a summary of the whole story from AutoCrit, a competing product that I'm trying out. They each have their strengths and weaknesses, but I do like the responsiveness of AC. It doesn't seem to have a Docs plugin, though. Regardless:

View attachment 2294314


It's trying to give a statistical analysis of... well, you can read the option I selected. It highlights "strong writing" as a weakness, so let's zoom in on that. I'm not going to hit all the subcategories, but here's the one that drags me down the most, along with a contrasting one:

View attachment 2294316

In most other cases, I'm doing "better" than the average romance story (as determined by AC), or at least on par. But that "filler word" thing is killing me! So why don't I fix it?

Because I write a lot of dialog, and that's how people talk. And my narrator is using a conversational style, and that's how people talk.

If I drill down, this is the summary it gives of the filler words I use "too much."

View attachment 2294318

Now, I did chop some out earlier where it was right. I had used "that" a few too many times where it could have been exempted when discussing action. The count was closer to 300 before that. But here are some examples that it wants me to remove:

View attachment 2294321

Could I take the "thens" out there? Sure. Do I think it would make it stronger? No, but AutoCrit does. I could make it read more like my English teacher would like, too, but it wouldn't read like real people trying to navigate a difficult situation. It would read like a computer's idea of an English teacher's idea of people trying to navigate a difficult situation, at best. But my summary score would be better!

This is what I mean. I'm using the tool, relying on it to help me make decisions. But they're my decisions. I'm not saying, "yup, take every one of these," because then it becomes a same-y mush of text. And I don't want that. And Laurel doesn't want that, either.

Is there a chance I'll get dinged eventually? Sure. Do I think the system is overly aggressive now? I have no idea. Without seeing writing samples, I'd still have no idea. But I also think that any attempts that are being made are in good faith. Laurel and Manu aren't pulling stories because they're jerks; they're doing it because they're worried. And I get why.

 

[NoTalentHack]

Fuck, I see another edit I want to make now, and it's one neither software suggested.

 

[MillieDynamite]

Dude, AutoCrit gives suggestions it doesn't rewrite. They are usually good suggestions but it's a tool to help you rework the work yourself. Your writer has an incredibly bad score on generic words and a low score on Pace and Momentum. He should pay attention to what it's telling him, and you should have him rework where necessary. @NoTalentHack, just my opinion.

I keep my gun near me, in case I get scores like that and feel the need to end it all.

I started using PWA June 5th. Since then, I've posted 14 stories and 1 essay, none of which encountered any issues in the queue, and two of which came in (probably, it's hard to be certain) fourth or fifth in contests. When I say that I take its suggestions "sometimes..." Hang on, let me give you an example.

This is from a story I'm editing now, in a section I've already gone over multiple times; I edit sometimes when I don't have the energy to write, so the early parts of my stories tend to be a lot "cleaner" than the later bits. Regardless, any additional changes are likely to be minor, and I've taken about all the suggestions from PWA I'm likely to take.

View attachment 2294304

The blue squiggly underline is Google Docs' sole (and incorrect) contribution. The blue underlines are from PWA, and they represent places where it thinks I've screwed up grammar or punctuation; PWA is either incorrect in these places or IDGAF. The yellow lines are places where it thinks I can improve my writing in some way: use of adverbs instead of stronger phrasing, extraneous words ("different possible" is one of the choices it flagged in this case, since they either could be used; that's probably true, but I like it better that way), passive voice, "glue" words or phrases ("even" or "just" as an intensifier, or "for that matter" as a stylistic choice), and so on.

Like I said, I take, in a 10K document maybe 10-15 of these suggestions and swap them out for something else. But the rest? I let them ride. I do use some other tools, too; this is the bare minimum of what PWA can use. But what I don't do is this:

View attachment 2294308


View attachment 2294310

I like my stupid, lazy phrasing. "Very obviously" is more conversational, which is... well, it's my voice. That's how I think, and it's how I write. I'm not saying someone could look at something I wrote and go, "Yup, that's an NTH story," but I think the people that follow me could read one of my stories and say, "Wait, this is an NTH story?" if I did something like that. Especially if I made a big shift to that.

If someone is using PWA and taking every one of those yellow underlines and saying "rephrase this," or even tinkering with them on their own before submitting until there are no yellow underlines? Yeah, they're going to have a voice that's recognizably not theirs.

For the curious, here's a mess of writing near the bottom where I haven't had a chance to edit at all. I have to get it all done by tomorrow! God help me.

View attachment 2294311

There are LOTS of things wrong with this section, but not that many more things that PWA caught. The red underlines are misspellings or words it doesn't recognize. The purple is a passive phrase it thinks it can rewrite; I may or may not change that, because I kind of like the passive voice there.

But there's stuff it hasn't caught, too: repetition of words, description of action that could be more vivid (because it recognizes adverbs as being signs of "weak" verb choice, but it sees lack of detail as clean, rather than boring), screwed up character indicators (there are four characters in the scene, but it doesn't "know" that), and so on. That's the other thing: these tools know what they know, but they don't know what they don't know.

This is a summary of the whole story from AutoCrit, a competing product that I'm trying out. They each have their strengths and weaknesses, but I do like the responsiveness of AC. It doesn't seem to have a Docs plugin, though. Regardless:

View attachment 2294314


It's trying to give a statistical analysis of... well, you can read the option I selected. It highlights "strong writing" as a weakness, so let's zoom in on that. I'm not going to hit all the subcategories, but here's the one that drags me down the most, along with a contrasting one:

View attachment 2294316

In most other cases, I'm doing "better" than the average romance story (as determined by AC), or at least on par. But that "filler word" thing is killing me! So why don't I fix it?

Because I write a lot of dialog, and that's how people talk. And my narrator is using a conversational style, and that's how people talk.

If I drill down, this is the summary it gives of the filler words I use "too much."

View attachment 2294318

Now, I did chop some out earlier where it was right. I had used "that" a few too many times where it could have been exempted when discussing action. The count was closer to 300 before that. But here are some examples that it wants me to remove:

View attachment 2294321

Could I take the "thens" out there? Sure. Do I think it would make it stronger? No, but AutoCrit does. I could make it read more like my English teacher would like, too, but it wouldn't read like real people trying to navigate a difficult situation. It would read like a computer's idea of an English teacher's idea of people trying to navigate a difficult situation, at best. But my summary score would be better!

This is what I mean. I'm using the tool, relying on it to help me make decisions. But they're my decisions. I'm not saying, "yup, take every one of these," because then it becomes a same-y mush of text. And I don't want that. And Laurel doesn't want that, either.

Is there a chance I'll get dinged eventually? Sure. Do I think the system is overly aggressive now? I have no idea. Without seeing writing samples, I'd still have no idea. But I also think that any attempts that are being made are in good faith. Laurel and Manu aren't pulling stories because they're jerks; they're doing it because they're worried. And I get why.

 

[NoTalentHack]

Dude, AutoCrit gives suggestions it doesn't rewrite. They are usually good suggestions but it's a tool to help you rework the work yourself. Your writer has an incredibly bad score on generic words and a low score on Pace and Momentum. He should pay attention to what it's telling him, and you should have him rework where necessary. @NoTalentHack, just my opinion.

I keep my gun near me, in case I get scores like that and feel the need to end it all.

LOL that writer is me.