[procone]

I found this really hard to read due to the Claude-isms. "Classic" chicken and egg problem? 39 em-dashes, random numbered lists, etc.

If you're not going to even bother to take the time to write an article, why should I waste my time reading it?

[Jtarii]

Started reading, saw "That was it, That was the whole reason", closed the page.

[RickS]

Is that a claudeism? IMO that's a perfectly natural trope that'd be at home in my voice or any of a million generic blog people I've read since long before AI. It is a linguistic trope, sure, but that's an unrelated criticism.

[Sparkle-san]

The thing with all AI-isms is that they all started humans, but we're apparently not allowed to use them anymore. Emily Dickinson obsessed with em-dashes, now they belong to ChatGPT.

[ruszki]

I’ve never read comments, articles, books in LLM’s current style before LLMs, and somehow, since then, I read comments like these every other day. This phenomenon is way overblown at the minimum. And since nobody showed a counter example from before 2022 which can be confused as LLM generated (using em dashes was never in it self that) I tend to say that it never even existed.

[slopinthebag]

You can use em dashes as long as you don't write like a bot.

[slopinthebag]

It's more subtle than that. Human written articles are immediately recognisable and so you let your guard down and tropes like this are accepted. But AI writing occupies a part of the uncanny valley where the hairs on the back of your neck stand up a bit, and every AI-ism like this is like a branch snapping beneath your feet

[Jtarii]

There is just a very specific kind of enthusiastic/authoritative/nerd voice that claude and other LLMs do does that is so grating to me. Like it's trying to be super upbeat and engaging and it comes off as trying too hard and is just offputting. If people are going to use LLMs to write their articles for them I would rather they be more neutral in tone, it just feels so incincere to get a robot to try to impersonate a enthusiastic human.

[hmcamp]

I read it and found it useful

\_(^ ^)_/

[542458]

I mainly object to AI writing when it’s excessively verbose. This was pretty information dense, a few AI-isms didn’t make it a waste of my time to read.

[pityJuke]

I've begun to find the writing style nauseating regardless of use, but yes, at least it isn't used to pad length.

[slopinthebag]

Yeah, I saw an oddly placed rule-of-three and closed the tab.

This website is going downhill, almost every blog post I open is just blatantly AI generated these days. It's like people don't have any self-awareness anymore, they just lazily prompt their AI to generate a post and then pat themselves on the back lol.

[debo_]

Welcome to Hell, Reader.

(I liked the article, personally.)

[Throaway199999]

pretty obviously a copy-and-paste job

[akomtu]

"How a Broken Bike Sync Led Me to Reverse Engineering My Wahoo's Hidden Debug Mode" - this is brain-dead AI slop right in the title.