its way too good at using the markers of having something to say, the sorts of things a professor might use after a period of lengthy explanation to make you snap back to attention and listen because this; This right here is the really juicy bit: but theres nothing there. Its using it to talk about whatever mundane thing you asked it to, with perfect neutrality and no substance at all.
I refer you to one of my favorite hacker news comments, I keep coming back to it and sending it to people:
I'd like to gently push back because this is a misconception that is worth untangling. It's not that no one is talking about it. It's that there's a genuine silence about the subject.
I feel the opposite. AI writing is like a version of Google Maps where you can see little black and white houses up close, but when you zoom out, all those details fade to white noise.
> Everything I read these days may have been written by Claude, so I’m constantly trying to determine if what I’m reading is or isn’t
I think it's even worse: There are also people who did write the original posts themselves, but then used AI to "polish" the style - i.e. reformulate a perfectly legitimate post as slop.
So even if you're reading obvious slop, you can't be sure if it's actual slop or a human writer "pretending" to be AI (because they haven't got the memo yet that is style is now a negative signifier, not a positive one)
I wonder if all of the fruits of LLMs come from the added training to build on top of a base or foundation LLM but also lead to all of the prose overfitting that we characterize as models having particular writing styles, rather than seeing a wider distribution of styles in response to prompts, in order to produce meaningful work.
its way too good at using the markers of having something to say, the sorts of things a professor might use after a period of lengthy explanation to make you snap back to attention and listen because this; This right here is the really juicy bit: but theres nothing there. Its using it to talk about whatever mundane thing you asked it to, with perfect neutrality and no substance at all.
I refer you to one of my favorite hacker news comments, I keep coming back to it and sending it to people:
https://news.ycombinator.com/item?id=48352444
Thanks for sharing that comment, it's excellent. I love the use of "midwife" in the last sentence. Claude would never!
It will now...
Ironically, the more polished AI writing becomes, the easier it is for me to recognize it.
Not because of grammar, but because everything feels just a little too intentional.
And honestly? Thats the part no one is talking about.
:|
I'd like to gently push back because this is a misconception that is worth untangling. It's not that no one is talking about it. It's that there's a genuine silence about the subject.
Let's pause for a moment, that actually sharpens the case rather than complicating it.
[flagged]
I feel the opposite. AI writing is like a version of Google Maps where you can see little black and white houses up close, but when you zoom out, all those details fade to white noise.
I really like the graph paper background. Is there a way to use this while having the letters fill exactly one block?
> Everything I read these days may have been written by Claude, so I’m constantly trying to determine if what I’m reading is or isn’t
I think it's even worse: There are also people who did write the original posts themselves, but then used AI to "polish" the style - i.e. reformulate a perfectly legitimate post as slop.
So even if you're reading obvious slop, you can't be sure if it's actual slop or a human writer "pretending" to be AI (because they haven't got the memo yet that is style is now a negative signifier, not a positive one)
> I hear Claude’s voice in a house, I hear Claude’s voice with a mouse.
The new schizophrenia.
Every time I call it out, I get downvoted here. But it legitimately has markers of a psychological illness.
Letting AI write for you is the main purpose though.
It should always have been trained on "is this more readable for humans" but as a actual tested thing.
I wonder if all of the fruits of LLMs come from the added training to build on top of a base or foundation LLM but also lead to all of the prose overfitting that we characterize as models having particular writing styles, rather than seeing a wider distribution of styles in response to prompts, in order to produce meaningful work.