logoalt Hacker News

torben-friisyesterday at 7:22 PM4 repliesview on HN

>Scalability, measured (the honest section)

Ugh.


Replies

mahoganyyesterday at 10:21 PM

The entire PR description is filled with LLMisms. I find this style so hard to read, it’s almost nauseating these days. My eyes start to glaze over and I stop reading pretty quickly after. I’m having to deal with this a lot at work these days unfortunately. I mean, did anyone even read or verify this output? When I read this type of stuff, I have the nagging suspicion that I am the proofreader and I have to do the verification myself. At that point, what are we even doing? I can generate Claude output myself…

I know I’m being overly dramatic but this sort of thing feels so wrong and inhuman(?) to me for some reason.

rbonvalltoday at 12:34 AM

Yes, that's the smoking gun!

show 1 reply
poly2ityesterday at 8:51 PM

I can't stand Claude's "honesty". Anthropic should hire some writers and linguists to make the output a bit more bearable. It's mentally taxing to read this type of dull text for hours every day.

show 1 reply
greenchairyesterday at 7:50 PM

almost spit out my drink!