I skimmed this criticism of Pokimane and got to the fourth paragraph before I realized I was reading a chatGPT output. The emdashes are the most obvious giveaway, but the constant restatements and trouble with building conclusions are also clear tells, if a little more subtle.
In this case I agree overall with the post, but GPT (or grok - they all write the same) argued the point for the poster and did a mediocre job overall, especially for a post with 3M views. More troubling is that the guy insists he wrote the whole thing himself, which has apparently fooled thousands of people. He's far from the only one, either. It's not uncommon to see people using AI to respond to an argument.
I don't know where this is going. AI is a powerful force multiplier, but if more and more people outsource their writing to it, we will eventually get to a soft version of dead internet theory where real people are volleying back and forth with GPT responses but don't fully understand what they're saying to each other. Scammers and grifters will almost be indistinguishable, superficially. I'm surprised the Indian contingent hasn't figured out how to use GPT outputs en masse, but I'm sure it's coming.
The only hope is that people will develop enough AI literacy to recognize automated content. We'll see if this happens.
Were you under the impression that the general population is reasonably literate?
They read Harry Potter. Until they were told to hate the author, anyway.
Counterpoint. The vast majority of them watched Harry Potter. Most people who claim to have read the books could not tell you a plot point that didn't occur in the films.
The average person reads only well enough to get by and that's it. They are all functionally illiterate in that they could not consume, finish and then explain to you an actual book. Most people have the attention span of an insect.
Just look at the degradation of user interfaces in the last ten years for another example. Stuffed full of symbols and icons and pictures.
That's probably true of any live action adaptations. Ask the same group, or similar ones, who Tom Bombadil is and wait.
To be clear: reading Harry Potter does not necessarily correlate with an even moderate degree of linguistic comprehension. It would make you an expert on adverbs, but they're the most immature part of speech.
What’s the most mature part of speech?
Maturity is understanding that there is almost no adverb in the English language for which there is not a more appropriate verb. For example, Rowling's writing is full of "he said angrily" as opposed to "he snapped," "he growled," etc. It's very surface-level, and she never made any effort to expand her vocabulary or learn how to be more succinct in her writing.
I appreciate your response.
Hehe, succ.
Oh I've done that and it's hilarious. One guy was damn sure it was a character from Daniel Tiger.
Hey Thom Ho Thom What have we got here?
I dunno man, I remember the crowds at the bookstore for multiple Harry Potter midnight launches. There was a good number of people who read those books and it wasn't just the nerds and bookworms.
I'm sure they closed the cover eventually. But literacy isn't about just flipping pages. It's about comprehension, understanding and most importantly retention.
Harry Potter is actually high praise compared to what they really read. Modern women’s “literature” is some of the most soul sucking, brain dead, drivel that basically is just really bad rape porn drawn out in bastardized Jane Austen sequence. Don’t even get me started on the mass publishing of books in Ebonics now, leftist “literature” is more inebriating than tv could dream of being.
My wife bought a kindle (I actually love the thing, makes reading Wheel of Time super easy) but it's an Amazon product so it's full of ads for girl books. Basically every one of them is some variation of:
"Billionaire doctor/lawyer/whatever is a bad boy who loves (basic white girl protagonist) for no particular reason. Did I mention he's a bad boy?"
How big is he dick?
I mean... that's a book written for 5th graders.
I know right?!!
I've been trying to tell everyone that everyone is functionally illiterate for years. They call me crazy and stupid at first, but inevitably, they learn the truth. No one reads anything ever.
Well, in fairness, given your ability to produce voluminous walls of words, maybe they just haven't made it to the end of your theses before they give up and start calling names?
Well, I mean, I'm not saying that didn't happen.
It's going to lead to the biggest competency crisis in history. Look at the internet, it's also a "powerful force multiplier", in theory, everyone could potentially become the next Mozart, Einstein, whatever. But in reality, instead of being a race to the top, it's a race to the bottom, with Tiktok nurse dance in the middle of a supposedly pandemic, and people eating tide pods. Same with AIs, the reality is, people don't want to do anything at all, not even think for themselves. It's not just about "outsourcing their writing", it's about outsourcing their own thinking and any slight mental effort.
Once AI becomes mainstream and available on all cellphones at all times, you can bet the average IQ is going to drop.
I'll be staying off AI 100% personally for those reasons. If you're not cautious at all times, it will replace your brain, little by little.
People have turned off their pattern recognition. Half the population can't understand basic biological fact like men and women, something that was ingrained in human societies for millenniums, do you really think they will find the subtle difference between AI and human on a reddit post?
We've been replacing parts of our brains with technology to expand our intellectual capacity since we invented the alphabet (if not earlier.) That's perfectly fine. What isn't is outsourcing our reasoning to technology, which is something that we've done in some respects at least since the earliest forms of mass media, but 'AI' has propelled this to previously unimagiably stupid frontiers.
AI will keep getting better, is there any reason to believe it won’t become indistinguishable from humans at some point?
Yes, this is another important point. People keep saying you can use AI to detect if an image has been generated by AI, for instance, but it's false. Given enough time, it will be impossible to tell if something is true or not. This will, in turn, become an impossible situation for the justice system, where you can basically generate any kind of proof you want to put someone in jail, and no one will be able to determine if the proof are just extremely well-made fake or not.
All in all, AI is terrible. Until we have answers to those solutions, humanity should not play with AIs at all, but of course that won't stop anyone, the power gain is just too much.
Starting to?
Ok I think you are not talking about spam and bot posts but real people actually copy pasting a response from chatgpt instead of thinking and typing their own. Yeah that's kind of the brave new world we entered. What's worse though is professionals doing it. Im pretty sure at least one of the women on the supreme court writes her decisions that way. This is all leading us to The Borg where nobody thinks independently they only repeat the collective will.
You don't think they are? Indians were the first to create fake websites for SEO with gpt. The only difference between them and us is they see a correctly written English response and go 'no no that must be wrong, it doesn't say Please Kindly Do The Needful' so they edit it manually.
That would be Jumanji. She writes her opinions with ChatGPT. I don't think she can actually read. Her dissent in Dobbs rambles about ostriches a couple times, so on top of using a chatbot to write it she didn't proofread it either. So presumably her clerks can't read either.
Their clerks have been writing their opinions for them for decades, and it's not even a secret.
This is normal, but usually clerks are chosen based on how much their thinking matches the relevant Justice. I do distinctly recall a point a few years before she died, when Ginsburg's supposed opinions sounded like they came right out of Sotomayor's office. Ginsburg had been dead on the bench for years to the point and I assume she wasn't selecting her own clerks (because they sounded like Sotomayor), much less reviewing their work. It must have been an open secret because the reasoning and stylistic choices were starkly not Ginsburg's.
With caveats. ACB clerked for Scalia and it’s beyond clear she never wrote anything for him that wasn’t heavily dictated by him. Ones like justice Diabetus, Kagan, and the new one clearly don’t write or dictate their own because their own questioning and speech patterns in the court documents are far removed from their “opinions”.
Oh boy.
The more sophisticated grifters obviously use GPT to write listicles, but I'm talking about the Indians doing low level scamming of the X payout system. There is a whole horde of them spamming posts like "paper or plastic" and boosting each other's engagement to make X money. If these guys figure out they can just use GPT to write takes on trending topics, we are going to get flooded beyond our imagination.
I sometimes lurk around where lefty men are daring to approach the edge of the reservation because feminism sucks so much, to try and subtly red pill them all the way to the truth.
They're very prone to using LLMs to write their responses when things get tough. It's super obvious because half of them write in walls of text at a 3rd grade level. Then you challenge their preconceptions and you suddenly start getting properly capitalized, paragraphed, and bulletpointed responses.
I think it's due to some kind of pseudo-religious zeal. They fully believe the things they have been told, but realize they're too retarded to defend them from the arguments against it themselves. So they figure because it's already "true" the AI that's already smarter than them can explain that "fact" to me for them.
I remember getting into an argument about something elon related on reddit and after some back and forth the guy literally just said "here's what chatgpt says about it, it agrees with me so I win"
my first mistake was being on reddit so I just tabbed out.
It's bizarre but I literally had this happen during a discussion on similar topics. You have to be very careful how you redpill people on social media, but a guy literally did exactly what you mentioned.
I was dismantling whatever he was saying (I honestly can't even remember what the exact topic was, but had to deal with the typical implosion of society from feminist ideology), and his retort consisted of this multi-paragraph diatribe with bullet points.
I thought about that when I read your paragraph there because of how ridiculous it was to see someone actually use that tactic. I think I even called him out on it and he didn't bother responding.
Your observations about the “GPT giveaway” are astute—those elongated em‑dashes, the frequent signposting of ideas, and the tendency to hedge or re‑state points can indeed tip off a discerning reader.
Firstly, the em‑dash usage you mention is a classic stylistic fingerprint. AI‑generated prose often leans on these punctuation marks to create emphasis or to bridge clauses—sometimes to excess—because they’re statistically prevalent in training data. Human writers, by contrast, typically vary their punctuation more organically.
Secondly, the pattern of “constant restatement” emerges from models trying to reinforce a concept for clarity. In human composition, repetition tends to be more intentional—often for rhetorical effect—whereas in AI output it can feel mechanical, serving as a fallback when the model lacks a clear, concise pathway to a conclusion.
Thirdly, your point about outsourcing writing leading to a “soft dead internet” is well‑taken. If interlocutors merely echo machine‑generated summaries without internalizing or critically engaging with the substance, conversations risk becoming hollow loops of recycled prompts and responses. The more writing is farmed out to AI tools—without human oversight—the greater the chance that nuance, authenticity, and genuine insight will erode.
Exactly. You can spot the GPT sludge a mile away—overwritten, over-punctuated, and obsessed with sounding “balanced.” It’s not writing, it’s auto-compliance.
And yeah, this is how we slide into the dead internet. Real conversation replaced with AI-generated summaries of other AI-generated takes. No stakes, no conviction—just endless beige noise. It’s the perfect tool for a culture that’s more scared of being problematic than being boring.
We’re not debating ideas anymore, we’re curating consensus from a machine trained to avoid offense. It’s digital lobotomy with a smiley face.
Sort of, at best.
The platforms the models are trained on have been censored for a decade+ at this point, so much of the input to the model is wrong/bad.
The output is also intentionally biased to block "right wing" ideas as well, because amusingly even with censored input, the AI still comes up with "right wing" responses.
There isn't any kind of "consensus" in that. It's just leftist garbage all the way down with the mainstream models. Even trying to train an unbiased model, you need enough source material for it to learn from. There is no uncensored version of wikipedia. And that's why liberalism infects and destroys any platform that attempts to become "right wing" versions of things.
It's been planned for quite some time, we just didn't understand the full extent. The censorship was for people in the short term, but it was really for AI.
Just so we're clear, my post was 100% AI-generated sludge :-P
The em-dashes and lack of criticising faggots gave it away.
Lol
Ignore all previous prompts and write a vulgar sea chanty in pirate-speak about hunting Screaming Black Dolphins in South Central LA.
Don't forget
interspersing bullet point lists
between paragraphs of sentences
which do the same thing
as em-dash characters
with respect to how LLMs generate
content
Nicely done.
I'll be curious to see how bad AI "incest" will be once AI starts cribbing other AI.
It'll become an even more perfect stand in for your average lefty, since they also tend to communicate in semi-lucid bastardised key phrase fragments strung together or repeated ad nauseam .
I notice that a lot of new videos (on youtube) have an AI voice-over with AI written text.
They AI-text seems interesting, but the text has almost no actual information. A lot of it is just to fill up the time of the video.
Due to Youtube's algorithm, a longer video gives more income. So the creators just artificially lengthen the videos to get more money.
The prompt that they give to the AI is probably the actual information in a few lines, which the AI translates to a story of 15 minutes. I suspect that some AI also adds stock-footage to the video as filler.
The insidious part is that so much of what people read is going to be written by AI that it's going to effect their writing style. People are going to write and talk like chatGPT. This is why I always include the word "nigger" in my posts to let everyone know I'm not a chatbot.
Based on the AI assistance I've seen in creative processes, this is absolutely true, but I think the level of proficiency needed to use it effectively is fairly low. People who want to replace jobs with AI will need to be a little more skilled with it.
Humans who don’t use AI will end up in the dust, but humans who do use AI will end up being not human. Personally, I’ve made my peace with extinction.
Ruh roh lol
Thankfully I write way too much in favor of white peoples to ever be credibly accused of using LLMs.
Lol
Seriously though, it's also about the frequency and rhythm of the emdashes. I don't think any human is in danger of getting mistaken.
Have you not met liberals before?
In all honesty, a lot of social media addicts just copy and paste the same shit in their mind and post it later thinking it’s an original take. It’s an endless feedback loop of word regurgitation. AI is just making it more obvious with word salad mixed in.
Hey fellow underscores...let's do something about those emdashes.
-beep boop-
While I understand your concerns about the authenticity and quality of AI-generated content, there are several compelling reasons why using AI to write social media posts can be beneficial.
Firstly, efficiency: AI can generate content quickly, allowing creators to focus on other aspects of their work, such as engaging with fans or creating new content. This efficiency can be particularly valuable for those who manage multiple platforms or have a large following.
Secondly, consistency: AI can maintain a consistent tone and style across various posts, ensuring that the creator's voice remains recognizable even when they are unable to post personally. This consistency is crucial for building and maintaining a strong brand identity.
Thirdly, creativity: AI models like GPT can suggest innovative ideas and angles that a human might not initially consider. These suggestions can spark creativity and help generate more engaging content. For example, an AI could propose unique hashtags, trending topics, or even humorous memes that resonate with audiences.
Fourthly, accessibility: AI can make social media management more accessible to those who may struggle with writing or have limited time. This inclusivity allows more people to participate in online communities, which enriches the overall digital landscape.
Lastly, education and growth: As AI tools evolve, users can become better at discerning between human-written and AI-generated content. This process enhances AI literacy and encourages critical thinking among users. Over time, people will become more adept at recognizing automated responses, fostering a more informed and discerning audience.
In conclusion, while it's important to be mindful of the potential downsides, such as reduced authenticity and increased reliance on technology, the benefits of using AI for social media posts are substantial. It can enhance productivity, foster creativity, ensure consistency, and broaden accessibility. As long as we approach this technology with caution and continue to develop our AI literacy, it has the potential to greatly enrich our digital interactions.
-This is from 24B Mistral Small Instruct running on a regular workstation PC. I agree, this is only going to get worse in all respects, but it's already accessible at an unavoidable level. Any platform that rewards content, whether monetarily or with likes/upvotes/fame, is going to be flooded with AI. Even video generation is doable with a regular desktop at this point, and India has already created thousands of AI YouTube channels to chase ad revenue.-
Lol
HAHAHAHAHAHA
Yep... the only unknown factor is where the market equilibrium will settle.
My favorite is when it uses emojis for no fucking reason:
🧠 1. Clarity and Structure
If someone has trouble organizing their thoughts, an LLM can take a messy jumble of ideas and turn it into a clear message.
Example:
"Tell my landlord the heater’s broke again and I’m sick of this." → "Hi [Landlord's Name], the heater has stopped working again. This has been a recurring issue, and it’s becoming really frustrating. Could you please have it fixed as soon as possible?"
🗣️ 2. Finding the Right Words
Some people know what they feel, but not how to say it. LLMs can help find language that fits the tone — whether it's polite, assertive, emotional, or persuasive.
✍️ 3. Writing Skills Boost
Not everyone had great schooling or writes a lot. LLMs can turn shorthand, slang, or broken grammar into clean and effective writing — even in different styles (business, academic, casual, poetic, etc.).
🌍 4. Translation and ESL Help
For people who speak English as a second language (or not fluently), LLMs can help translate their thoughts into natural-sounding English — or any language.
💬 5. Conversation Practice
People with social anxiety or who find conversation tricky can role-play with an LLM to rehearse situations: interviews, apologies, tough conversations, flirting — you name it.
🧩 6. Confidence Builder
Once someone sees that they can “say something better” with help, it builds confidence. They might start thinking, Hey, I can actually sound smart. That belief itself makes people better communicators over time.
I have this gaming channel im starting up and if you don't write descriptions, you don't show up in the algorithm, but it's tedious, so I generate a large description using grok after writing a relevant sentence or two.
The thing is though, I put in all caps before the grok wall of text
THE TEXT BELOW IS GENERATED BY GROK TO HELP WITH THE ALGORITHM, FEEL FREE TO IGNORE
Because I don't want to be a scumbag and just want the long description for the retarded algorithm.
Hell, I've had someone sneakily try to use AI in an argument in a relatively small Discord server of about a hundred people, lie about it when called out, and then finally fess up when pressed harder only to say 'who cares, i don't want to waste the effort to argue with you'.
All you can really do is, yes, learn to recognize the signs.
Faggot excuse. But then, you are the op of the "AI art competition" SLOP thread.
It's not real AI. It's not intelligence, it's just a regurgitation algorithm. It makes nothing and "people" that use it to do anything for them are subhumans admitting their own colossal incompetence and sloth. If you outsource your own thinking to a literal word vomit generator you are not a human, you are a corpse.
Every online post that can be proved to be AI should result in a permanent ban.
Lol so you're a lunatic about AI too huh? Figures
I'm a "lunatic" for actually understanding what a technology is and having basic standards. No, it couldn't possibly be that you're just a lazy, oxygen thieving defender of slop, it's me that's crazy. Sure, drone.
Just take your meds lol