>Long, grammatically perfect comments that sound hollow and a bit lengthy

It's worse than I thought. They've already managed to mimick the median HN user perfectly!

No problem, I'm here to keep the language sophistication level low.

I tried to make ChatGPT generate a counterpoint for that but it turns out you're right.

Yes. The old heuristics of if something is generated by grammar and sentence structure don't work as well anymore. The thing that fucks me up the most about it is that I now constantly have to be uncertain about whether something is human or not. Of course, you've always had to be careful about misinformation on the internet, but this raises the scalability of false, hollow, and harmful output to new levels. Especially if it's a topic I'm trying to learn about by reading random articles (or comments), there isn't much of a frame of reference to what's good info and what's hallucinated garbage.

I fear that at some point the anonymity that made the internet great in the first place will be destroyed by this.

I'm a non native English speaker. Edge new feature of automatically improving my text is a God send. Unfortunately it is blocked at work.

Many business doesn't want to send their data to a third party such as OpenAI, so until locally run LLM becomes wildly available in businesses.

as the meme goes "always has been"

i remember seeing the change when GPT-2 was announced