Do you have links to any examples? I can’t say I’ve seen a whole lot of what you’re describing apart from a few isolated posts or comments here or there. I’d suggest checking to see if a significant portion of these posts are coming from a specific community or from a specific instance, and then adjusting your blocks to accommodate.

P.S. Since we’re on the topic, I’m not a bot; I just talk and write like this.

P.P.S. That’s not entirely true. I would have written “I’m not a bot—I just talk and write like this.” but it didn’t seem like an appropriate time to use an em dash.

Edit: after looking through OP’s profile and seeing a couple of apparently AI-generated posts, I’m starting to think that this post is also AI-generated or that OP has some weird false flag-esque agendum that they’re up to

Edit 2: the banner picture on their profile literally reads “LLM❤️”. They are a troll spam account.

Hi Mudkip, you are totally right! Our posts are getting very sterile. We will try to delve into the reason for this and do a better job going forward.

Nobody talks like that. Nobody. […] Or when someone uses “delve” unironically. No real person has ever said delve.

There’s literally dozens of us. Dozens!

Nobody talks like that. Nobody

How do you think LLMs learned to talk like that? Plenty of people talk (or rather, write) exactly like that. It’s just now heavily associated with AI because AI has been specifically tuned to output sentences structured that way.

Yes, AI has a vibe to it that can be very predictable and easy to spot, but that doesn’t mean every single instance of someone knowing how to string a few paragraphs together with certain verbiage is a bot.

Idk maybe I’m being paranoid

Seems that way. Healthy skepticism is good and necessary. Paranoia not so much.

This attitude is essentially discouraging certain syntactic styles in writing. I and many others now regularly have moments when writing posts where we go, “Hm, I maybe shouldn’t word it this way or people might say I’m a bot,” and it’s obnoxious that it’s come to that.

And the comments are worse!! Someone will ask a simple question and get back three paragraphs of perfectly structured nothing that somehow says less than “idk man” would have.

You would have loved Usenet in the 90s. That’s how a lot of people are.

Sometimes I say delve. Like “I hate delves” when I played world of Warcraft. Which I don’t anymore.

I write really long posts and I’ve said “delve” exactly four times in my Lemmy career. I know this because I just checked.

I’m like 99.8%, maybe 99.9% certain I’m not an AI.

As an avid enthusiast of Lemmy, I’d like to delve into this topic a bit…

Nah, can’t do any more of it.

I can’t say I’ve seen a particular uptick in slop, but I’ve seen other red flags. Ah well.

Well, some people naturally have a flair in their writing style. * thinks back to the times when asterisk talk was cool * * sighs discretely to avoid attracting attention *

I say “delve” often in technical breakdowns when writing. I never say it out loud because my hands will tell you we’re digging down into a topic. I use em dashes too, when digressing into an important fact that needs to be mentioned but not expanded upon. It comes from textbooks, not plagiarised LLM gable. Call me assumptive, but I’d expect ley has a higher rate of similar-minded people than other platforms - people that have properly studied textbooks at some point.

You can probably tell I’m human because I usuallyiss a typo somewhere in my post where I hit backspace instead of “M” at the lead of a word, causing two words to questionably mash together. Autocorrect can’t help with those. But that’s what an LLM would say if it knew it’s secrecy was in jeopardy

I think you might be getting false positives on LLM ID, but you’re justified in being skeptical. The Slop is everywhere, no reason to think we’re hidden

Just do yourself a favor and block the obvious bots or report them if they’re straying outside of the rules.

I noticed it a lot in “news” feeds, which I quickly blocked because I don’t want to fucking doomscroll, but they often ARE bots mindlessly mirroring “news” articles that are increasingly being generated by shitbots.

After I blocked all the bullshit, Lemmy/Piefed is actually pretty good, if a touch quiet.

Though I know how infuriating it must appear, below the surface is much more complicated and nuanced than it first seems. Perhaps if I threw in a paragraph outlining why or how people notice that maybe an ai was writing the sentence, it’d become clear.

But I’ve seen enough hentai in my life to know how this turns out.

idk man

I don’t think this is happening but I’ll be on the lookout

To truly delve into the art of writing is to embrace the challenge of constraints. I will endeavor to delve deeply into my lexicon, striving to use the word delve precisely nineteen times. As I delve into this task, I must delve past the initial awkwardness and delve into a rhythm that makes the word feel natural, not forced. My intention is to delve beyond a simple word count exercise and delve into the craft itself. One could delve into the history of the word, or delve into its etymology, but I will not delve into such topics here. Instead, I will delve deeper into the immediate requirement. I must delve into the nuances of sentence structure to accommodate this repetition. To delve this way is a unique experience. I delve into my creativity to ensure each use works. As I delve further, the task becomes clearer. I can delve into the flow of writing with ease. This final sentence will delve my intent home, ensuring I have met the demand and can delve no more into this specific requirement.