Hacker News Reader

Comment by kevinh | Hacker News Reader

This article is like 300 words. Would it have killed them to not generate it using AI?

Maybe this will change one day but at the current moment this is an immediate turnoff. It's like someone trying to show you their project day 1 and it's a page filled with ads and a newsletter popup. You may have good reasons to do that but it doesn't instill a sense of trust and quality.

I don't know how much of it was hand-edited and how much was direct output, but this article has that unmistakable LLM voice. The rhythm, the rhetorical flourishes; it's all there even if it's diffused through some human revision.

The really weird thing is going to be when people start internalizing the LLM voice and writing that way. It's probably happening already.

I've seen many people do the latter, I get quite annoyed by it. Worst of all is wondering if I'm affected by it myself, I doubt most people who've gotten an 'LLM writing style' know so themselves.

Eventually no space where people can just 'publish' things will be safe from being completely filled with LLM writing/video/images. The only way to combat it is by forcing people to get punished for this behaviour and making it difficult to circumvent.

Some invite system where people get punished for the bad people they bring in, one that's linked to your identity/workplace/education. Even if these options were available, I doubt many people would care enough, they'd rather be in 'enshittified ' spaces.

I'm so embarrassed to say that I read it and didn't notice. But now that you pointed it out, I reread it and you are so right. It is clearly generated.

I have flagged this article on principle. Idk if it it's in the spirit of HN to do that or not, but the button's there, and I'm going to use it.

This actually seem quite flag-worthy to me. Look at the rest of the site, it's not at all trustworthy. The first post says it's by some random 16 year old (if we can actually believe that) and only has a few posts. One of them is a comparison of smart watches which says they tested them in the subheading on the article listing, but then doesn't show anything more than a surface level comparison from AI.

Reply on news.ycombinator.com