• The regular administrative staff are taking a vacation, and in the meantime, Biigoh is taking over. See here for more information.
  • A notice about Rule 3 regarding sites hosting pirated/unauthorized content has been made. Please see here for details.
  • Staff is working to deal with the problem of synonymous tags. See here for more information and to suggest tag mergers.
  • Due to issues with external spam filters, QQ is currently unable to send any mail to Microsoft E-mail addresses. This includes any account at live.com, hotmail.com or msn.com. Signing up to the forum with one of these addresses will result in your verification E-mail never arriving. For best results, please use a different E-mail provider for your QQ address.
  • For prospective new members, a word of warning: don't use common names like Dennis, Simon, or Kenny if you decide to create an account. Spammers have used them all before you and gotten those names flagged in the anti-spam databases. Your account registration will be rejected because of it.
  • Since it has happened MULTIPLE times now, I want to be very clear about this. You do not get to abandon an account and create a new one. You do not get to pass an account to someone else and create a new one. If you do so anyway, you will be banned for creating sockpuppets.
  • Due to the actions of particularly persistent spammers and trolls, we will be banning disposable email addresses from today onward.

Penalty for not disclosing ai work

00loner

Depress , motivated ( may be )
Joined
Feb 11, 2020
Messages
4,972
Likes received
16,225
Theres many fics which doesn't has ai either in the tags or in description but is actually ai fics. Even reading a few paragraphs, you will understand that this is ai slop. There should be some kinds of penalty for this purposeful deception.

Although some people disclose the ai part in either tag or in description or in both, some intentionally hid it and despite asking them to clarify it they put their head in the sand and ignore the requests and suggestions.
 
They have many times said that tag enforcement will not happen. Stop trying to bring about AI witch hunts. Even if you personally aren't trying to, that is the singular and inevitable conclusion to what you are calling for.
So I am just supposed to waste my previous time reading ai slop when I can read some authentic work?

Also, witch hunt is both bad and sounds bad. I am taking about enforcing rules so that people don't waste their time.
 
So I am just supposed to waste my previous time reading ai slop when I can read some authentic work?

Also, witch hunt is both bad and sounds bad. I am taking about enforcing rules so that people don't waste their time.
It's not a rule precisely because there's no reasonable way to enforce it. We can talk all about how it's "obvious" when someone is using AI, but at the end of the day there's no way to actually prove that something was written by AI.
 
Spend a bit of time looking into the threads all about the exact topic you are talking about and see exactly what you seem to be lacking in knowledge of. What you are asking for has a 100% failure rate for it everywhere.
Besides, the mods of all levels have full on, flat out said they will NEVER EVER do tag enforcement. Period, full stop, no questions asked, end of statement.
 
Any rule suggestion which requires multiple moderators to proactively and analytically read many tens of thousands of words per story we likely don't care about, and then discuss amongst ourselves, and then determine a punishment structure, just so a single tag for one thread can be applied, is a total nonstarter for us. It is just not something we are going to logistically allow, because we volunteer moderators would rather spend those hours of our lives playing video games or whatnot instead.

Beyond that, we've said many times we're not planning to enforce any tagging standards, because that's a major hassle of the same vein. A process which would require a responding moderator to read entire hundred-thousand word stories for any given instance, just to develop a personal assessment which may be prone to error, simply cannot become an expected routine. Tag application and description will remain in the hands of the authors as a general rule.

Ultimately, we're all adults here – you can make your own quality judgements by reading a bit of the story, irrespective of from whence the literary flaws derive.

(Also see the previous thread for more discussion on the topic: https://forum.questionablequesting.com/threads/making-ai-tags-mandatory.36944/)
 
Theres many fics which doesn't has ai either in the tags or in description but is actually ai fics. Even reading a few paragraphs, you will understand that this is ai slop. There should be some kinds of penalty for this purposeful deception.

Although some people disclose the ai part in either tag or in description or in both, some intentionally hid it and despite asking them to clarify it they put their head in the sand and ignore the requests and suggestions.
If you can tell its AI, then you don't need the tag. If you can't tell it's AI, then you don't have a problem.
 
Is there not dome kind of plugin that can flag things as AI? As there are sites like GPTzero that can do so with high accuracy.
If you can tell its AI, then you don't need the tag. If you can't tell it's AI, then you don't have a problem.
The problem is that most AI written stories seem to have the first few chapters edited manually, to be actually readable. As you get further, it turns less into AI assisted and more into AI written. Stories like that clutter my searches, and require me to read past the first few edited chapters to start being noticeable.
 
Last edited:
Is there not dome kind of plugin that can flag things as AI? As there are sites like GPTzero that can do so with high accuracy.

The problem is that most AI written stories seem to have the first few chapters edited manually, to be actually readable. As you get further, it turns less into AI assisted and more into AI written. Stories like that clutter my searches, and require me to read past the first few edited chapters to start being noticeable.
They don't do it with any degree of accuracy actually. Most AI detectors will flag works written long before AI even existed as AI since that's what the AI is copying from. Any work written by a human with any sort actual writing background or professionally edited will fail those.

AI detectors are only even mildly useful for essays and such, not stories, and even then they're still very prone to throwing false positives.
 
Would a public poll on top of the story in question work, with AI, not AI, and unknown?

This way the readers can be the judge instead of taking too much admin time. Just wait till 100+ votes orso. Combined with some way of automated testing like the GPTzero, would what I believe to give together a fairly accurate indicator. While not even requiring to read a word.

And if still wrongly judged, then an admin could be asked to take a look.
 
Would a public poll on top of the story in question work, with AI, not AI, and unknown?

This way the readers can be the judge instead of taking too much admin time. Just wait till 100+ votes orso. Combined with some way of automated testing like the GPTzero, would what I believe to give together a fairly accurate indicator. While not even requiring to read a word.

This works until I go in and vote a new story as being AI without reading it and get a few of my friends to do the same; while for larger more popular stories my single vote would be washed away as meaningless, for a new writer posting thier first story having 100% AI on the poll would turn away new readers quick.

Why would I do so? Maybe I'm a troll. Maybe I had an argument with them in the politics forum. Maybe I had an argument with them in space battles. Maybe they criticized my work. Maybe they used a tag I don't like. Maybe they didn't use a tag I do like. Maybe I'm just having a bad day and want to spread it around. Maybe I'm just a troll. Whatever the reason, now this new story is voted as being AI, and will have a lot harder time getting readership.

This should demonstrate why such a poll will not be implemented.
 
Gptzero mentioned itt is actually the only ai detector that has reasonable free limit and doesn't detect classics as ai, so use that and block the author.

Also since it's the Nth thread on the topic I'd like to propose relaxation of "don't be a dick" rule in threads that have patreon links only: I should be able to say "this smells like ai and these detectors agree with me". Because attempting to sell ai generated writing on a writing forum is the next level scumbaggery.
 
AI Overview




In literary criticism, the "colors" of prose refer to descriptive slang used to classify writing styles based on their density, tone, and pacing. It is a metaphorical color wheel that helps writers, editors, and readers critique how language serves—or distracts from—the story. [1, 2, 3, 4]
A quick breakdown of the most common prose styles includes:

💜 Purple Prose
  • What it is: Writing that is overly flowery, ornate, and melodramatic. It often sacrifices plot and pacing for indulgent, showy descriptions.
  • Key traits: Excessive adjectives, adverbs, run-on sentences, and an over-reliance on a thesaurus.
  • Vibe: "Like showing up in stilettos to go on a hike". [1, 2, 3]

🤎 Beige / Tan Prose
  • What it is: The exact opposite of purple prose. It is plain, stripped-down, and highly minimalist.
  • Key traits: Short sentences, basic vocabulary, and a flat, objective, or neutral tone.
  • Vibe: Direct, straightforward, and to the point—often found in journalism or hard-boiled detective fiction. [1, 2, 3, 4]

💙 Blue Prose (Language)
  • What it is: Language that is sexually explicit, crude, or profanity-laced.
  • Key traits: Heavy use of vulgarity or "talking a blue streak" (a colloquialism for talking fast, at length, or with expletives). [1, 2]

💛 Yellow / White / Pink Prose
  • Yellow: Factual and simple prose that focuses entirely on the sequence of events without adding emotional layers.
  • White: Completely objective, invisible writing where the author's voice is stripped away entirely so the reader only sees the story.
  • Pink: Light, fluffy, and highly romanticized writing. [1, 2]


To explore this further, if you'd like, you could:
  • Tell me what genre you are writing (e.g., fantasy, sci-fi, thriller) to help determine which style fits best.
  • Provide a sample paragraph of your own writing so we can evaluate its "color."
Let me know how you'd like to proceed with your writing!
This is pain google.

In any case you don't even need an AI to tell if a story is AI ghostwritten. At some point the style trips over itself due to memory limits. But new writers these days may use AI-Assistance to learn how to write never noticing how it colors their own stories. Some of these later chapters may be hand-written by using lessons from how the AI edits it. Which the AI takes an inch and turns into a mile.

As a result there is no way to prove something is entirely AI-generated. Humans all too easily pick up bad habits. For me when a story reads like a series of talking heads that's bad. And such a writing style has existed since before AI, they just picked up on it and have no context to realize it's bad. That's the problem with mass-training models.


Rather than "it reads like AI" some criticism towards the style of the work might be better? Like the fact some NSFW stories are clinical in their sexual parts. In that they exist but no NSFW scenes are forthcoming, ever. Like AI they endless loop around themselves and never get to the meat you come for in such stories. But that's not an AI problem, but the data they were trained on.
 
And, so that you do not have this rather minor irritation, you'd open up everyone else to witch hunts.
It would certainly be easier with a poll but fake reporting with a group of friends can already be a thing, you just need to use the report button currently.

It would limit the mods to only look into Not-AI claims, as opposed to AI ones. If there would be consequences, to claiming not AI after a poll that could stop everyone from claiming it.

Maybe the poll only appears after being reported as AI a few times.
Some of these later chapters may be hand-written by using lessons from how the AI edits it. Which the AI takes an inch and turns into a mile
I mostly stumble across stories which are written seemingly normally, before completely switching to AI. It might just sound more normal with all polishing done to the first few chapters.

My main argument for public voting after reports for being undisclosed ai would be the same as that of the moderators: AI allows for the generation of many different stories a day, reading through each one to check for AI suck. Moreso when they clutter search results.

Even ignoring stories and creators isn't truly an option when new accounts can be used to bring more ai content. It could be sockpuppets. (Which is against the rules, but I wouldn't be able to tell which are sock puppets either with them being all AI).
 
Last edited:
Even ignoring stories and creators isn't truly an option when new accounts can be used to bring more ai content. It could be sockpuppets. (Which is against the rules, but I wouldn't be able to tell which are sock puppets either with them being all AI).
Thankfully those can be tracked via IPs. Sockpuppets get caught and all the linked accounts get banned.
 

Users who are viewing this thread

Back
Top