Forum: Story Recommendations

The problem with all of these is they're completely predictable. You know how it's going to end before you finish reading the first few paragraphs.

And yes, I've read all of most of them. Got too bored with a couple to finish.

I can, because they're chock full of AI tells.

Please, share your insights to the exact 'tells' you see.

Thanks.

It's been about 48 hours and AJ has not come back to share his experiences.

Can anyone else that has read Megumi Kashuahara's tales that are tagged AI offer their experiences with noticing the 'tells'?

Well, unrealistic "overpowered" main characters, nothing can harm the MC, usually no character development, particularly because the entire story takes place in a very short amount of time, and nothing much happens, and the MC has a solution or response to everything happening... that kind of things are a strong sign...

I do not say that apply to all of Megumi's work, but the feeling is there for the one I read, even though I liked it. But I did not enjoy it so much that I would remember the title.

I haven't read the story in question, but in my limited experiments with AI writing there's some things I would consider AI tells:

- hive mind: characters are unexpectedly aware of everything, with or without any explanation how. Like someone being perfectly informed of details of conversation that just happened in different place, instantly. Also including characters that can randomly say lines or pick up treads that do belong to them.

- repetitiveness: this can get as bad as several paragraphs repeated in a cycle, with or without slight variation, but that is easily edited (if the author care to do any editing at all), Worse are the small things like constructs like "[character] made as sound like [nonsense]" used with the same character and different nonsensical descriptions increasingly frequently throughout a scene, and/or the same signature moves of characters repeated beyond where it makes any sense. This also includes cyclical dialogue where characters escalatory prize each other and hype up "let's do it" or similar while making not an real inch toward the goal, as well as repetitive scenes and/or repetitive scene triggers.

- drift: from generally poor continuity to derailed details, like if there's at all any description of clothing or the room or setting in general, it gets often retrofitted if not entirely confused, and changes/state are not tracked creating continuity errors (like supposedly removed clothing that reappear on).

- forced deconfliction: just about any tension is resolved asap and in most efficient way possible in complete disregard to characters ability or even willingness to do so, including freely breaking continuity to do so, or introducing new, unexpected elements. Lying doesn't exist or if occur is not tolerated to an extreme degree for anything over a paragraph or so in length (remember the hive mind, everyone knows everything anyway).

- phrasing and constructs... well, now were in the em-dash territory. Nothing of this is definitive (neither are anything of the above), but things like "it's not [this] but [something else]" especially when there's poor match between the parts. Nonsensical similes, especially when part of repetitive patterns.

Yeah, also paragraphs composed of a sentence stating the negation of something abstract, then a positive statement of some other abstract term, either related or not, but usually not the opposite of the thing negated in the first sentence of the paragraph, e.g.:

Not .... The ... (something very loosely connected, so you have to rack your brain how it even fits into the picture).

dialog for several people in the same paragraph

I've seen this in archived (and dead-tree) pulps from the 50's/60's, so it's not a new thing. If done well, it's transparent, and a lot less intrusive than the multiple single-short-line paragraphs that would otherwise be needed (and uses less paper).

For an example of how bad multi-speaker well-tagged dialog can be, see this transcript of Jasper Carrott's hit B-side

Our thoughts on this story, like the stars in the heavens, are aligned. I couldn't make it past the second chapter.

So this is what sent me to the forums. I took a look at the recent top stories/serials pages and they are chocked full of ai generated works. Given how people dislike ai junk, I'm wondering if I was seeing voter fraud, bots cheering on bots.

So this is what sent me to the forums. I took a look at the recent top stories/serials pages and they are chocked full of ai generated works. Given how people dislike ai junk, I'm wondering if I was seeing voter fraud, bots cheering on bots.

The tags mean people who dislike AI written stuff will avoid these stories, so only people voting are those who don't care or actually like the sort of garbage most AI written stories consist of.

The ratings reflect the views of people who actually read the stories, the tags ensure those are (mostly) people who like stories with those kinds of tags.

I'm glad you said 'mostly'. I've read some that I really liked and I've even recommended a couple.

That means you'd fit in the group I mentioned without the qualifier. The qualifier was because of the occasional idiot who doesn't read the tags and then complains about a story of a type they don't like, or jerks who downvote a story because it had tags they don't like.

Some authors who use AI generation in their stories avoid the AI-generated tag and their stories get stellar scores, so clearly the reader bias against AI generation isn't based on the stories themselves.

First - if the avoid the tag how do you know they use AI generation?
Second - unfortunately there isn't currently a way to separate "used AI as part of the tools to create this story and then worked to make it a good story" and "got AI to generate this slop without doing any work at all", not at least without reading more of the story than I care to generally. The dislike of AI generation is because the flood of the later.
Third - I'd appreciate links to these stories

The threads which address AI detection (including this one) list a number of AI tells. If you get too many in a story, it's a strong sign AI-generation was used in whole or part.

Two problems with this. First, the "AI tells" are also "bad writing tells". If the story has too many of those then it is not, by definition, a well written story.
Second, there are plenty of people who have those AI tells in their stories, long before there were any AI.

Off topic, but a colon can completely change a sentence.
Example:
Susie licked Amy's ice cream cone.
Susie licked Amy's colon.

And a semi-colon requires medical treatment. :)

Sorry I have no connection to him/her. I've been following the discussion here on AI use and came across their stories and found them interesting. I see she won a Clitorides so some others must agree. No idea how much of her writing is AI.

Oh, and are you suggesting I'm AI generated? Not at all, although Mrs. Rodeodoc still seems to think my tallywhacker is a real machine.