onehitwanda
Venatrix Lacrimosal
- Joined
- May 20, 2013
- Posts
- 3,807
kitty nibbles are one of my love languages.
The AI Rejection Conversation Matters
- Thread starter scbolder
- Start date
StillStunned
Writing...
- Joined
- Jun 4, 2023
- Posts
- 6,912
That sounds like a synonym thrown up by AI. "You've overused 'eating pussy'. Try this instead." Fuck off, AI. There's no such this as too much eating pussy.
onehitwanda
Venatrix Lacrimosal
- Joined
- May 20, 2013
- Posts
- 3,807
By about hour number three things are getting a little tired down there.
AwkwardlySet
On-Duty Critic
- Joined
- Jul 24, 2022
- Posts
- 3,077
Wow, I thought the story count was somewhere around 100k, maybe 200k, but this is a truly overwhelming number. It gives even more power to the point I have been making lately.
StillStunned
Writing...
- Joined
- Jun 4, 2023
- Posts
- 6,912
Sounds like you'd need a .... wait for it ... you know what's coming ... come on, say it with me ... a cat nap! Hahahaha my wife's on the other side of the world and I haven't interacted with anyone for a week.
onehitwanda
Venatrix Lacrimosal
- Joined
- May 20, 2013
- Posts
- 3,807
jaF0
Watcher
- Joined
- Dec 31, 2009
- Posts
- 38,556
It seems like the policy has been reworded. Can't be sure because I didn't keep a copy of whatever it was before and there is no changelog, buty there are sections in there I don't remember seeing.
https://www.literotica.com/faq/publishing/publishing-ai
https://www.literotica.com/faq/publishing/publishing-ai
Britva415
"Alabaster," my ass
- Joined
- Nov 19, 2022
- Posts
- 3,077
This made me look at past versions with Wayback Machine.
Interesting - ever since the very beginning, that page has told people to come to AH and discuss it.
The "What Is Considered an AI Generated Story at Literotica?" section was added last December.
Portly_Penguin
Experienced
- Joined
- Apr 29, 2023
- Posts
- 78
I'm not around here much any more so I wouldn't know
AwkwardMD
Belzebutts
- Joined
- Apr 13, 2014
- Posts
- 1,944
Any time a work is rejected, for any reason, subsequent submissions of the same work take anywhere from twice to ten times the normal submission time to get a response, good or bad.
Thanatos_X13
Pervert
- Joined
- Dec 16, 2008
- Posts
- 50
I think it was like 12 days for me.
Five_Inch_Heels
Unexpected
- Joined
- Nov 28, 2015
- Posts
- 1,437
And mine were approved literally within a few hours of submission and published the next morning. There is something in common that some of you are doing or using to trigger the bot.
That could be because you have more stories published already. If you get hit with the AI rejection notices after your first submissions like me, its possible they flag your account internally or something. Then you're stuck trying to climb out from the hole with the odds stacked against you.
Five_Inch_Heels
Unexpected
- Joined
- Nov 28, 2015
- Posts
- 1,437
I didn't have any at all and the first one went up within two days.
There is some wording in the policy that suggests editing tools rewriting paragraphs could be part of the issue.
There is some wording in the policy that suggests editing tools rewriting paragraphs could be part of the issue.
My first one went up in 2 days as well, and its written like my other stories. For the actually writing, I only use tools to correct grammar and spelling and then send it off to my editor (who also seems to have no idea why my stories are getting flagged). I'm wondering if there's some part of my writing style matches the patterns that some AI models output, and that's resulting in it being incorrectly flagged.
djrip
Oneirographer
- Joined
- Apr 28, 2015
- Posts
- 1,827
Ask a dozen AI detectors about a given piece, and you'll get a dozen different results. Might as well throw darts at post-its on the wall with 'yes' and 'no' written on them. Maybe that's what lit is doing?
The first few times I was able to improve my stories a fair bit after the rejection. But you very quickly run out things to fix, and having to wait 3 weeks or more until it gets accepted or rejected makes it hard to troubleshoot what wording is causing the issue.
I logged on today and saw that 3 of my 4 pending submissions rejected for AI, with the 4th submission now showing up as 'new'.
I'm happy that one finally made it through, but I'm sort of at a loss right now as to why those 3 submissions were rejected. They either score 0% or less than 4% on ZeroGPT, and there's nothing obvious that stands out to me.
I'm happy that one finally made it through, but I'm sort of at a loss right now as to why those 3 submissions were rejected. They either score 0% or less than 4% on ZeroGPT, and there's nothing obvious that stands out to me.
Last edited:
As I said I don’t believe ANYTHING the sites Admins put out as purported “data” is accurate. If they cannot make a public statement about their AI witch hunt backed by hard evidence then I choose to not believe any of this sites data. I will never “submit” a story to an undocumented process.
Having seen possibly one of the best new authors in the last decade hounded off this site by continual false AI rejected I have nothing but contempt and disgust for the admins of this site.
Having seen possibly one of the best new authors in the last decade hounded off this site by continual false AI rejected I have nothing but contempt and disgust for the admins of this site.
cadeauxxx
la conteur
- Joined
- May 13, 2015
- Posts
- 181
As someone who returned to these forums after years of not being active, all due to the false AI detections, I want to give my 2 cents and advice for anyone that needs help.
I had 3 stories rejected at the end of last year. Having not used AI or knew much about it at the time, I was very upset and offended that this happened to me. Like so many authors this happened to, I took to these forums to complain and screech about how I was being mistreated. Most people feel like they're being accused of something, so it's completely understandable if you're reacting defensively.
Over the year, I've done a lot of research to understand AI and how the detections are working and why they make life hell on us. No one on this forum knows what exact tools that the editors are using for detections. That's out of our knowledge, but it don't matter. There are ways we can prevent it.
One of my rejections (from December 2023) overused the same adjectives multiple times. I was quite embarrassed by this, since it made me realize I did a poor job proofreading the story while being sick with COVID. So in reality, the rejection helped me realize a problem on my end and go around and fix it by rewriting multiple sentences and deploying a thesaurus to pick newer, better words to describe something.
After that incident with my last story, I have not had another issue with approving stories.
Since my last visit on this forum, I have learned that there are websites that generate text with AI.
These generators are built on LLMs (Large Language Models) that compile hundreds and thousands of texts that it copies from documents online. I believe they copy entire websites too, but I can't prove that. The LLM builds an entire language based off the texts that it gets, and the result is that it generates predictive texts with stock phrases and builds on them. These tend to be generic texts yet read coherently from the AI.
I believe the false detections are happening due to repeated usage of specific words (mainly adjectives) in describing things in sentences. It can be as simple as using the same adjective several times throughout a large body of text, or even paragraphs.
It sounds rude to be told "write better", kinda like "get good" in gaming terms. But that really is the solution.
The best advice I can give is to use a thesaurus. There are multiple websites like Collins Dictionary and Thesaurus.com that are a great help and you might even find words you've never used before that you can include to spice up your vocabulary.
We tend to overlook when we are repeatedly using the same words. You can copy/paste your story here at Wordcounter and the little bar on the right side will show you the most overused words under Keyword Density. From there, you can see what's going on in your writing and carefully edit it.
Hope this helps someone.
I had 3 stories rejected at the end of last year. Having not used AI or knew much about it at the time, I was very upset and offended that this happened to me. Like so many authors this happened to, I took to these forums to complain and screech about how I was being mistreated. Most people feel like they're being accused of something, so it's completely understandable if you're reacting defensively.
Over the year, I've done a lot of research to understand AI and how the detections are working and why they make life hell on us. No one on this forum knows what exact tools that the editors are using for detections. That's out of our knowledge, but it don't matter. There are ways we can prevent it.
One of my rejections (from December 2023) overused the same adjectives multiple times. I was quite embarrassed by this, since it made me realize I did a poor job proofreading the story while being sick with COVID. So in reality, the rejection helped me realize a problem on my end and go around and fix it by rewriting multiple sentences and deploying a thesaurus to pick newer, better words to describe something.
After that incident with my last story, I have not had another issue with approving stories.
Since my last visit on this forum, I have learned that there are websites that generate text with AI.
These generators are built on LLMs (Large Language Models) that compile hundreds and thousands of texts that it copies from documents online. I believe they copy entire websites too, but I can't prove that. The LLM builds an entire language based off the texts that it gets, and the result is that it generates predictive texts with stock phrases and builds on them. These tend to be generic texts yet read coherently from the AI.
I believe the false detections are happening due to repeated usage of specific words (mainly adjectives) in describing things in sentences. It can be as simple as using the same adjective several times throughout a large body of text, or even paragraphs.
It sounds rude to be told "write better", kinda like "get good" in gaming terms. But that really is the solution.
The best advice I can give is to use a thesaurus. There are multiple websites like Collins Dictionary and Thesaurus.com that are a great help and you might even find words you've never used before that you can include to spice up your vocabulary.
We tend to overlook when we are repeatedly using the same words. You can copy/paste your story here at Wordcounter and the little bar on the right side will show you the most overused words under Keyword Density. From there, you can see what's going on in your writing and carefully edit it.
Hope this helps someone.
I currently have 5 stories pending at the moment (for over 2 weeks now), and in all cases I've made sure that they scored 0% on www.zerogpt.com. So hopefully they aren't rejected this time.
cadeauxxx
la conteur
- Joined
- May 13, 2015
- Posts
- 181
Best of luck to you.
Be patient with it pending. Sometimes it takes a long time. Don't resubmit, or you'll be waiting even longer.
Its not just words that detectors look for. There are also a multitude of different AI text generation models available, some of them only accessible through websites and others available for download.
AI models are trained from human content, and thus they learn patterns that are present in human content. This is why people end up getting falsely flagged for AI, because their writing style contains the patterns that the detector AI noticed in the text generation AI's outputs.
AI detectors are looking for patterns commonly associated with the outputs of certain AI models. These patterns can take the form of words, but they can also be the burstiness of your writing (ex: how much sentence length varies) and the perplexity (amount of information) contained in your text. AI detectors also try to look at content, and the temperature (an LLM setting that controls creativity) of the text.
Last edited:
