So we deleted everything. Every product, Every photo. Legally, that may change nothing. Emotionally, it did.
Both of these are pure LLMisms. If you told claude your story and asked it to make a infographic picture to share on social media you would get really similar writing.
I use AI a lot at work and have a hypersensitive sense for it. Everyone in this community is LLM-deaf because the majority of members don’t use it enough to recognize it’s default voice in something.
The top and bottom paragraph might be their own words its too short for me to tell.
It leads to hilarious situations like this where users unironically share LLM generated content (in part or in whole) and when I point it out I get downvote bombed. It’s doubly ironic when users shame me for calling it out to protect the AI…
So sometimes I’ll argue with users here with claude on default settings, specifically prompting it to be obvious that it’s claude but not too obvious. I’ve never been called on it.
There’s also this:
People who frequently use ChatGPT for writing tasks are accurate and robust detectors of AI-generated text (Russell et al., ACL 2025)
In Russel et al’s study just 5 “expert” users of AI could predict AI generated text with a majority vote. They had a 299/300 accuracy far exceeding existing AI detectors or the general public.
There are also studies showing if you don’t use AI a lot you don’t recognize it. Shocker…
again, you might be right, but your arguments are poor.
Perhaps you get this every single time because a statement of “this is AI” followed by “trust me bro” isn’t as convincing an argument as you seem to think it is ?
I don’t dispute your ability to detect LLM generated text in larger samples, but this sample is way too short and too “natural” to be sure enough for such an accusation. How about you ask them directly whether they generated this text or not, and take their word for it? Because what’s the point of accusing them of using an LLM when it’s 1) impossible to prove and 2) irrelevant for the world at large?
Good writers use more than just these few tricks where AI seems to struggle to not use them everywhere. Do you remember what posts like this would look like from some random etsy member pre-AI? Were they all expert writers using the same punchy voice?
Again… the unbelievable irony of being flamed for calling a post AI made on the fuck-AI community…
When you have no evidence other than “it uses effective writing techniques” of course you’re going to get flamed. You sound like a conspiracy theorist who thinks they can see something no one else can and it makes you special. Whereas all you’re really doing is distracting from the message of the post and trying to police everyone’s writing style.
Do you remember what posts like this would look like from some random etsy member pre-AI? Were they all expert writers using the same punchy voice?
Like youtube face/thumbnails? or listicles? or you know, the whole idea of clickbait.
Convergence to the most effective format of a media type is an established phenomena, especially in areas where there is some incentive.
again, you might be right, but your arguments are poor.
Perhaps you get this every single time because a statement of “this is AI” followed by “trust me bro” isn’t as convincing an argument as you seem to think it is ?
I used to have a ton of soap etsy listings and a large soap Instagram. I am super wordy.
I had competitors cropping up making repeated listings after ai was popularized using my tone and cadence in their listings. I used to be teased in the soap community cause I wouldn’t shut up because id get so excited about every little detail id include in each soap. LLM I guess thought it was great so now that’s what listings look like for so many other people. I had a tag line I would end all my posts with, stay sudsy, and people were even generating descriptions and including that too. I know I don’t own that phrase, but every post for 7 years ended in stay sudsy and I guess it got scraped from Instagram.
I had built a business and community over years and it was just snatched from me. And the best part was many of those ‘soapers’ found more success than I could because they churned out more bulk basic soap, but had the same character of my small boutique business.
LLMs overweight the style of this particular context of short explanatory posts (not quite long enough for a full blog post but generally longer than a Facebook status or a tweet) for public consumption. LinkedIn is where a lot of this type of post lives, and LLMs tend to borrow from those conventions and extend them outside of that context.
But that stuff started there for a reason — em dash clauses are an efficient method to give a broad statement and then elaborate/explain in detail after, or by giving an example after.
Bullet points are an easy way to visually organize things into lists
Bullet points also are easy to textually encode in markdown, which is a friendly format for both training data and LLM output, because of the way LLMs work
It’s not just that the “it’s not X, it’s Y” sentence structure comes from LLMs; rather, it becomes some kind of logical negative space, that complements the other type of em dash explainer — the general then specific elaboration.
Many of us learned to write in specific mechanical styles for corporate communications. Those rules aren’t that complex, so it stands to reason that those rules end up getting incorporated into how machines generate text, too. But most of us have been doing it for a while, and oscillate between avoiding sounding like LLMs and defiantly using those conventions and saying stuff like why should I change, he’s the one who sucks.
No, I didn’t quit using em dashes, bullets, or “it’s not X” sentences. But I do use them more judiciously. And because of that, I recognize when they are still absolutely worth the overhead of people accusing the text of being LLM-written.
Dude it’s a large language model, of course there will be intersections with normal speech patterns or people may speak that way themselves. Especially as the output of LLMs becomes more and more commonplace and intersects with our normal vernacular and cadence.
I get this every single time… there are specific tells for LLMs and y’all are blind to them I swear…
I would agree with you if it wasn’t structured like unedited vanilla LLM output that people meme about
It’s not X, it’s Y.
There’s also the plain as day pandemic of content creators defaulting to AI to write viral punchy posts because it can do it so quickly. Youtube is 80% AI generated at this point as well it seems… nobody writes their own scripts anymore.
You don’t have to believe me. I think my strongest points are the study I linked and this community’s inability to tell when I argue with them using LLMs. If anywhere on Lemmy is going to have AI content slip past detection it’s the community where people say fuck AI and try not to use it.
do you have a base for that accusation?
Both of these are pure LLMisms. If you told claude your story and asked it to make a infographic picture to share on social media you would get really similar writing.
I use AI a lot at work and have a hypersensitive sense for it. Everyone in this community is LLM-deaf because the majority of members don’t use it enough to recognize it’s default voice in something.
The top and bottom paragraph might be their own words its too short for me to tell.
It leads to hilarious situations like this where users unironically share LLM generated content (in part or in whole) and when I point it out I get downvote bombed. It’s doubly ironic when users shame me for calling it out to protect the AI…
So sometimes I’ll argue with users here with claude on default settings, specifically prompting it to be obvious that it’s claude but not too obvious. I’ve never been called on it.
There’s also this:
People who frequently use ChatGPT for writing tasks are accurate and robust detectors of AI-generated text (Russell et al., ACL 2025)
In Russel et al’s study just 5 “expert” users of AI could predict AI generated text with a majority vote. They had a 299/300 accuracy far exceeding existing AI detectors or the general public.
There are also studies showing if you don’t use AI a lot you don’t recognize it. Shocker…
I can only quote another poster who told you:
I don’t dispute your ability to detect LLM generated text in larger samples, but this sample is way too short and too “natural” to be sure enough for such an accusation. How about you ask them directly whether they generated this text or not, and take their word for it? Because what’s the point of accusing them of using an LLM when it’s 1) impossible to prove and 2) irrelevant for the world at large?
Fuck you for giving a monopoly on a writing style to AI. AI copied good writers and those people still exist.
Good writers use more than just these few tricks where AI seems to struggle to not use them everywhere. Do you remember what posts like this would look like from some random etsy member pre-AI? Were they all expert writers using the same punchy voice?
Again… the unbelievable irony of being flamed for calling a post AI made on the fuck-AI community…
When you have no evidence other than “it uses effective writing techniques” of course you’re going to get flamed. You sound like a conspiracy theorist who thinks they can see something no one else can and it makes you special. Whereas all you’re really doing is distracting from the message of the post and trying to police everyone’s writing style.
Like youtube face/thumbnails? or listicles? or you know, the whole idea of clickbait.
Convergence to the most effective format of a media type is an established phenomena, especially in areas where there is some incentive.
again, you might be right, but your arguments are poor.
Perhaps you get this every single time because a statement of “this is AI” followed by “trust me bro” isn’t as convincing an argument as you seem to think it is ?
I used to have a ton of soap etsy listings and a large soap Instagram. I am super wordy.
I had competitors cropping up making repeated listings after ai was popularized using my tone and cadence in their listings. I used to be teased in the soap community cause I wouldn’t shut up because id get so excited about every little detail id include in each soap. LLM I guess thought it was great so now that’s what listings look like for so many other people. I had a tag line I would end all my posts with, stay sudsy, and people were even generating descriptions and including that too. I know I don’t own that phrase, but every post for 7 years ended in stay sudsy and I guess it got scraped from Instagram.
I had built a business and community over years and it was just snatched from me. And the best part was many of those ‘soapers’ found more success than I could because they churned out more bulk basic soap, but had the same character of my small boutique business.
It sucks.
It’s a full shitshow and the rate at which shitty people can be shitty is the highest I’ve ever seen.
LLMs overweight the style of this particular context of short explanatory posts (not quite long enough for a full blog post but generally longer than a Facebook status or a tweet) for public consumption. LinkedIn is where a lot of this type of post lives, and LLMs tend to borrow from those conventions and extend them outside of that context.
But that stuff started there for a reason — em dash clauses are an efficient method to give a broad statement and then elaborate/explain in detail after, or by giving an example after.
It’s not just that the “it’s not X, it’s Y” sentence structure comes from LLMs; rather, it becomes some kind of logical negative space, that complements the other type of em dash explainer — the general then specific elaboration.
Many of us learned to write in specific mechanical styles for corporate communications. Those rules aren’t that complex, so it stands to reason that those rules end up getting incorporated into how machines generate text, too. But most of us have been doing it for a while, and oscillate between avoiding sounding like LLMs and defiantly using those conventions and saying stuff like why should I change, he’s the one who sucks.
No, I didn’t quit using em dashes, bullets, or “it’s not X” sentences. But I do use them more judiciously. And because of that, I recognize when they are still absolutely worth the overhead of people accusing the text of being LLM-written.
Dude it’s a large language model, of course there will be intersections with normal speech patterns or people may speak that way themselves. Especially as the output of LLMs becomes more and more commonplace and intersects with our normal vernacular and cadence.
Isn’t that just the nature of the beast?
I get this every single time… there are specific tells for LLMs and y’all are blind to them I swear…
I would agree with you if it wasn’t structured like unedited vanilla LLM output that people meme about
It’s not X, it’s Y.
There’s also the plain as day pandemic of content creators defaulting to AI to write viral punchy posts because it can do it so quickly. Youtube is 80% AI generated at this point as well it seems… nobody writes their own scripts anymore.
burden of proof is on the accuser.
Not to say you have to justify your every comment, but people asking for justification is to be expected.
a “trust me bro” doesn’t usually carry that much weight.
the LLM equivalent of :
You don’t have to believe me. I think my strongest points are the study I linked and this community’s inability to tell when I argue with them using LLMs. If anywhere on Lemmy is going to have AI content slip past detection it’s the community where people say fuck AI and try not to use it.
I was simply stating your strongest points were unconvincing in general type.
In this specific context, still unconvincing , but that doesn’t imply correctness, just lack of persuasive efficacy.