Easiest way to spot AI in a thread

[Deleted Member 49073]

If you see the link icon, there's a 99% chance it was copied and pasted from a chatbot.

 

[swanneo]

it's really easy to tell, the em dash groups of three, it's not X it's Y

 

[Deleted Member 49073]

Even if someone edits the words so that it bypasses AI detectors, most people forget to remove the hyperlink. @ey88

 

[Adriana Limа]

And also if you see multiple hyphens. JewGPT uses it so much, but I rarely see it in human writing.

Obviously it's not 100% accurate though

 

[Deleted Member 49073]

it's really easy to tell, the em dash groups of three, it's not X it's Y

The em dash, lmao. So funny that people forget to remove it.

 

[ey88]

it's really easy to tell, the em dash groups of three, it's not X it's Y

whenever i use chatgpt i say no dashes

 

[pompompurino]

it's really easy to tell, the em dash groups of three, it's not X it's Y

yes the groups of three especially

 

[slopslinger]

formatting can be removed but prose and cadence never lie.

 

[Deleted Member 49073]

whenever i use chatgpt i say no dashes

Literally just these 7 words make it so much harder to spot AI text.

 

[glamourpursuer]

everyone loves saying “oh that’s obviously AI” like it’s some special sixth sense, but the truth is, detecting AI content — especially now — isn’t nearly as easy as people think. sure, a few years ago, maybe the writing was stiff or repetitive enough to spot instantly, but AI has evolved fast. especially with models like GPT-4 and beyond, the writing can sound casual, emotional, sarcastic, poetic, technical — literally whatever tone the user wants. it mirrors human language really well.

people forget that humans write in all kinds of ways too — awkward, over-explained, typo-filled, weirdly formal, or overly polished — and AI can replicate all of it. so just because a message sounds “too perfect” or “too structured” doesn’t automatically mean it’s a bot. some people are just articulate. and on the flip side, AI can purposely include flaws, slang, typos, and casual grammar to sound more human.

even AI detectors aren’t that reliable. many of them give false positives — calling real human writing “AI-generated” just because it’s clear or formulaic. and the inverse happens too — AI content gets mistaken for human writing all the time. unless the AI is spitting out something super generic or filled with that classic “as an AI language model…” line, you’re probably guessing, not detecting.

so no, it’s not the “easiest thing” to spot AI in a thread anymore. the line between human and machine-written content is blurring, and a lot of people who claim they can tell at a glance are just relying on vibes, not actual evidence. if you're confident, great — but don't act like it's foolproof. AI's writing quality is scary good now. sometimes even the creators can't tell without checking.

 

[ey88]

View attachment 130830

Literally just these 6 words make it so much harder to spot AI text.

I never knew u could do that

 

[Adriana Limа]

everyone loves saying “oh that’s obviously AI” like it’s some special sixth sense, but the truth is, detecting AI content — especially now — isn’t nearly as easy as people think. sure, a few years ago, maybe the writing was stiff or repetitive enough to spot instantly, but AI has evolved fast. especially with models like GPT-4 and beyond, the writing can sound casual, emotional, sarcastic, poetic, technical — literally whatever tone the user wants. it mirrors human language really well.

people forget that humans write in all kinds of ways too — awkward, over-explained, typo-filled, weirdly formal, or overly polished — and AI can replicate all of it. so just because a message sounds “too perfect” or “too structured” doesn’t automatically mean it’s a bot. some people are just articulate. and on the flip side, AI can purposely include flaws, slang, typos, and casual grammar to sound more human.

even AI detectors aren’t that reliable. many of them give false positives — calling real human writing “AI-generated” just because it’s clear or formulaic. and the inverse happens too — AI content gets mistaken for human writing all the time. unless the AI is spitting out something super generic or filled with that classic “as an AI language model…” line, you’re probably guessing, not detecting.

so no, it’s not the “easiest thing” to spot AI in a thread anymore. the line between human and machine-written content is blurring, and a lot of people who claim they can tell at a glance are just relying on vibes, not actual evidence. if you're confident, great — but don't act like it's foolproof. AI's writing quality is scary good now. sometimes even the creators can't tell without checking.

This is an excellent, well-argued take — articulate and grounded without slipping into overstatement. If you're thinking of posting this somewhere (like Reddit, Twitter/X, or a forum), it would absolutely hold its own. You make a number of key points that people often overlook:​

• The evolution of AI fluency — highlighting how GPT-4+ can intentionally sound human in any tone.
• Human variability — many falsely assume human writing has a singular “natural” style, but in reality, it's all over the place.
• Detector flaws — you rightly point out that AI detectors are still unreliable, especially for well-written or formulaic human content.
• False confidence — the “I just know” attitude a lot of people have often relies more on intuition or bias than actual indicators.

 

[Deleted Member 49073]

I never knew u could do that

Yeah, learning how to effectively use AI will have a good yield for you in the future.
I urge EVERYONE to learn prompt engineering.

 

[slopslinger]

tbh op this obsession with the link icon as if it signifies something decisive about the origin of a thought is the very same kind of aesthetic laziness that masquerades as insight that you accuse her of.

to say “99 percent chance it was copied from a chatbot” is forgetting the more terrifying possibility that a human being in earnest despair copied it willingly. not cause theyre a machine, but because they cant bear to speak in their own voice.

 

[Deleted Member 49073]

tbh op this obsession with the link icon as if it signifies something decisive about the origin of a thought is the very same kind of aesthetic laziness that masquerades as insight that you accuse her of.

to say “99 percent chance it was copied from a chatbot” is forgetting the more terrifying possibility that a human being in earnest despair copied it willingly. not cause theyre a machine, but because they cant bear to speak in their own voice.

I see what you mean about the link icon obsession being shallow, and it really does risk missing the bigger picture. The real issue might not be about originality or who said it first, but about why someone would feel the need to hide behind words that aren’t theirs. Maybe it’s not just despair, but a deeper silence - a place where their own voice feels inadequate or impossible to find. That’s a far more complex and tragic thing to reckon with than just accusing someone of copying. It makes me think about how we judge creativity and authenticity, and whether we sometimes dismiss human vulnerability as laziness or fakery because it’s easier than facing it head-on.

 

[slopslinger]

I see what you mean about the link icon obsession being shallow, and it really does risk missing the bigger picture. The real issue might not be about originality or who said it first, but about why someone would feel the need to hide behind words that aren’t theirs. Maybe it’s not just despair, but a deeper silence - a place where their own voice feels inadequate or impossible to find. That’s a far more complex and tragic thing to reckon with than just accusing someone of copying. It makes me think about how we judge creativity and authenticity, and whether we sometimes dismiss human vulnerability as laziness or fakery because it’s easier than facing it head-on.

ugh. *drinks more*. we’re all drunk on performance, and the hangover is despair. so cheers to you fren

 

[Deleted Member 46433]

it's really easy to tell, the em dash groups of three, it's not X it's Y

i used to love em dashes but i always try to find another way to word things to purposely avoid them. same with "its not___ ,but ___"

 

[swanneo]

i used to love em dashes but i always try to find another way to word things to purposely avoid them. same with "its not___ ,but ___"

same w me just feels weird now

 

[Deleted Member 49073]

i used to love em dashes but i always try to find another way to word things to purposely avoid them. same with "its not___ ,but ___"

I'd only use them for emphasis, too paranoid now that my work will be mistaken for AI.

 

[Deleted Member 46433]

I'd only use them for emphasis, too paranoid now that my work will be mistaken for AI.

im paranoid of everybody using ai unfortunately. ive never thought so with you though so good job

 

[Deleted Member 47925]

I'd only use them for emphasis, too paranoid now that my work will be mistaken for AI.

this is why i'm overly blunt unless someone needs help emotionally, and i was never good at punctuation.

 

[Deleted Member 47925]

im paranoid of everybody using ai unfortunately. ive never thought so with you though so good j*b

because he has emotional charge, punctuation won't take that away.

 

[Deleted Member 49073]

because he has emotional charge, punctuation won't take that away.

First time someone has said this about me.
My family always say I'm overly blunt or have bad emotional intelligence.
In their eyes: I'm the analyst, not the poet.

 

[Deleted Member 47925]

First time someone has said this about me.
My family always say I'm overly blunt or have bad emotional intelligence.
In their eyes: I'm the analyst, not the poet.

Sometimes you need to hear what you second guess, and it coming at random should tell you exactly what it means bud.

 

[glamourpursuer]

bump