4 ms·

Can you please not post AI-generated or AI-edited comments to HN? It's not allowed here - see https://news.ycombinator.com/newsguidelines.html#generated https:/

by dang 5mo ago

Can you please not post AI-generated or AI-edited comments to HN? It's not allowed here - see https://news.ycombinator.com/newsguidelines.html#generated https://news.ycombinator.com/newsguidelines.html#generated and https://news.ycombinator.com/item?id=47340079 https://news.ycombinator.com/item?id=47340079. Of course, it's impossible to know for sure what was LLM processed, but your posts are getting classified that way and, on inspection, this does seem justified. We tend to ban new accounts that post like this, but I don't want to ban you because https://news.ycombinator.com/item?id=48310559 https://news.ycombinator.com/item?id=48310559 sounds like you have interesting contributions to make ("I work with vehicle specification data daily."), which is great! Just please write any text that you post to HN by hand.

plewd 5mo ago
I'm curious why this comment got flagged? At least to me it looks really human (typos, uncapitalized words, etc..)
- sperandeo 5mo ago
  im curious too because i am not using ai and now im literally scrutenizing how to even talk here. Even asking ai how this even looks ai? what is even going on here?
  - dang 5mo ago
    I responded to the GP comment because it was your most recent post, but there have been quite a few other posts that got classified as genai - e.g.: https://news.ycombinator.com/item?id=48310576 https://news.ycombinator.com/item?id=48310576 https://news.ycombinator.com/item?id=48310559 https://news.ycombinator.com/item?id=48310559 https://news.ycombinator.com/item?id=48310458 https://news.ycombinator.com/item?id=48310458 As I said, it's impossible to know for sure - but we have no choice but to use imperfect measures because (1) the site is getting inundated, and (2) the community feels strongly about it. In most cases when I've traced through similar cases with specific users, it turned out that they were using some tool (e.g. a grammar checker) that they didn't think of as AI-editing, but which was leaving LLM imprints on their text. So far it doesn't seem possible to run text through LLMs and not have that happen, but we have no idea how that will evolve over time. For anyone wondering, I suppose I should add that this is in no way anti-AI! LLMs are amazing and we rely on them heavily; we just don't use them on text that we post to HN, since HN conversation is a social activity for humans. There was an interesting PG post about this the other day: https://x.com/paulg/status/2058844147092488401 https://x.com/paulg/status/2058844147092488401, https://xcancel.com/paulg/status/2058844147092488401 https://xcancel.com/paulg/status/2058844147092488401 Although he was writing about emails, not forum comments, the argument is much the same.

by dang 5mo ago

plewd 5mo ago

I'm curious why this comment got flagged? At least to me it looks really human (typos, uncapitalized words, etc..)

sperandeo 5mo ago
im curious too because i am not using ai and now im literally scrutenizing how to even talk here. Even asking ai how this even looks ai? what is even going on here?
- dang 5mo ago
  I responded to the GP comment because it was your most recent post, but there have been quite a few other posts that got classified as genai - e.g.: https://news.ycombinator.com/item?id=48310576 https://news.ycombinator.com/item?id=48310576 https://news.ycombinator.com/item?id=48310559 https://news.ycombinator.com/item?id=48310559 https://news.ycombinator.com/item?id=48310458 https://news.ycombinator.com/item?id=48310458 As I said, it's impossible to know for sure - but we have no choice but to use imperfect measures because (1) the site is getting inundated, and (2) the community feels strongly about it. In most cases when I've traced through similar cases with specific users, it turned out that they were using some tool (e.g. a grammar checker) that they didn't think of as AI-editing, but which was leaving LLM imprints on their text. So far it doesn't seem possible to run text through LLMs and not have that happen, but we have no idea how that will evolve over time. For anyone wondering, I suppose I should add that this is in no way anti-AI! LLMs are amazing and we rely on them heavily; we just don't use them on text that we post to HN, since HN conversation is a social activity for humans. There was an interesting PG post about this the other day: https://x.com/paulg/status/2058844147092488401 https://x.com/paulg/status/2058844147092488401, https://xcancel.com/paulg/status/2058844147092488401 https://xcancel.com/paulg/status/2058844147092488401 Although he was writing about emails, not forum comments, the argument is much the same.