Comment by cortesoft

FWIW, I tried to read the AI version first, and found myself bouncing back to the prompt multiple times. The AI version just reads as flat, impersonal, and as if it has had all the corners sanded off (both in style and content). The prompt conveys much more about what _you_ want to say, and is easier to read without zoning out and losing interest.

If you must use AI in this way, I'd tell it to take a _much_ lighter touch. It cleans things up without washing away the personality of your writing. ChatGPT example:

> adjust the following for spelling and grammar, making MINIMAL alterations to the original wording and content. do NOT rewrite, just clean up what is already there: [post]

https://chatgpt.com/share/6a8382b9-3620-83eb-b17e-4397aee5ad...

Yes, the original prompt is 10000% better than the AI output, because it shows me exactly how much the author (you?) cares about this topic and how much effort they're willing to put into communication, which is barely any at all.

The stream-of-consciousness in the prompt is much better than the LLMs response. Yes, it's messy and unpolished, maybe even a little hard to parse, but it's so much richer for it. The way it is written lets me follow your thought process, shows that you were getting the words out quickly without much or any editing, and how you were exploring the idea and making it more concrete by giving it written form.

The LLM response has none of that character, it doesn't give me that little sliver of insight into the writer. It strips the idea from the author, making it generic and anonymous. The only thing I can gleam from text edited with an LLM like this (besides the literal meaning of the text), is that the author didn't care enough to express themselves.

Sure, it's more polished, but doing the writing manually is half the point. You have an abstract idea, you refine into something that can be communicated, and in doing so, explore that idea and improve your understanding of it. I've re-written this paragraph a dozen times trying to do exactly that.

I think you intentionally sandbagged your prompt by dragging on and being repetitive and it’s still a better read than the LLM output.

There’s also a difference between “please edit my writing” and “please generate a blog post from these bullet points.”

I asked claude 4.5 : """ Please, with an absolute minimum of content modification, edit this text with proper paragraph-formatting, spelling, punctuation and grammar. Do not alter vocabulary, language, or tone. """

And then fed it your input.

And the output was a very easily readable, enjoyable plain-ascii-text, with no Ai-ick, except for an em-dash or two.

Tried pasting it here, but that removed paragraph formatting

This is actually a great representation of what is good and what is insidious about LLM assisted writing.

The practice of writing can serve the purpose of conveying an idea or information to the readers, but it can also be an invaluable tool for the writer to process, digest, and internalize a message.

The act of wordsmithing, and shaping the idea from its raw rambling form into a coherent narrative, is intellectual metabolism.

A good writer will both distill the ideas, present them in such a way that will be pleasurable to their readers, and also bring the reader along the way so they can learn as well, and feel like an active participant.

The LLM tend to produce prose that defeats the purpose for the writer, takes the joy out for the reader, and ultimately does nothing for anyone but take up space.

It's not more words that make good stories, and it's not more code that makes good programs. It's what use is this to anyone, and only humans understand what other humans want and only humans can care.

Seems like people in this thread agree that it's better to express oneself in a less skilled fashion but actually write something that represents your idea, than to present some polished turd and pass it off as your own writing.

Same as a birthday card on some stock paper written in crayon will always be more valuable than a store bought card with generic sentiment.

The stream-of-consciousness is only hard to read because it lacks paragraph breaks, sometimes sentence breaks, and there's so many misspelled words. If you just tweak it to fix that stuff, it becomes fairly readable, like most internet comments or whatnot. (Which do you think this comment is, a polished essay, or lightly-formatted stream of consciousness?)

Meanwhile, the LLM-generated essay is hard to read because it's essentially the high-school essay that's padding its paper to meet a minimum page count. I'm surprised that you find it easier to read than most human text, because I've invariably found that my first reaction to LLM-generated text is "where's the point?"--everything is just buried in so much fluff.

So yeah, I find the prompt much better than the LLM output.

I think it's hilarious how AI took something personal and the very first thing it spews out is a classic AI-ism.

Do you read over what the AI spits out? Do you read over what you write?

It takes time to do either, and both are the things that I think are actually useful.

Sure enough, the prompt is infinitely more expressive and interesting.

AI writing is like the voice anonymizer they'd use in TV interviews when they wanted to hide the identity of the person speaking. The output is more adherant to convention but likewise more impersonal and exhausting, and when it's used by too many people it's so indistint that it becomes noise.

It might look good to you in isolation, because you're excited to see that it says what you meant to say without the features you feel self-conscious about, but to others it just reads like yet another sample of the same old AI slop. The message gets entirely lost behind that.

I read your entire jumbled stream joyfully and hated the llm article that was spit out.

definitely prefer the rambling; i can break it up into paragraphs sans LLM

while both are difficult to read, i enjoy reading the prompt more, it feels more human (don't know exactly what).

The author giving the slightest amount of shit seems like a big assumption to make, but ultimately even it that were the case I don't see why I would prefer what I read to be curated in that specific manner (having access only to what has been filtered past some black box). As opposed to being able to choose what to read from the whole chat history.

I don't like it. I don't take any offence. It just triggers the same AI;DR for me.

If I think the task requires a human to think about it. And the person is the human paid to think about it. Whether I get a wall of AI text, prompt, or an entire AI conversation... there is no human insight in it.

If that person took the time to think about the problem, had a chat with the AI, and understood the output, surely they can take the final step and tell me the short result/summary themselves, right?

---

However, if I had to choose lesser evil, I would of course prefer to have a link to the AI conversation over having a wall of text directly in the conversation.

They could prompt the AI to scroll up and copy their previous prompts for them.

Yeah, by the time my thought is fully formed I've gone through an entire conversation.

Even when LLMs add something it's a better idea to let the readers use LLMs to augment content. They have the best knowledge of their own background and what they're looking for.

Sharing information that is already in the corpus is simply not that valuable in a world where everyone has access to their own LLMs.

I'm in the "post both and let your readers figure out which, if any, they prefer" camp but even if you didn't (and unless you were using one of those super special models that can't be released according to the companies) I would hope that any literate adult that thought it valuable would be capable of consulting something with comparable knowledge and experience.

Of course, given that posting one is not mutually exclusive with posting the other, I would still think that posting both is the easiest way to satisfy both people who prefer to read the prompt and those who prefer the read the output.