[freehorse]

AI has a distinct, low-quality and unnecessarily sensational writing style, which signals low effort post first of all. If you use AI and you put the effort to make it not look like AI it means that at least you put some effort into it.

I am not sure how the future may be, but the direction of AIs has been to solidify their distinct writing style rather than assimilate to humans. So we can start addressing what we are dealing with now. I doubt HN tagging/flagging can put enough pressure to affect how llms will evolve anyway in the future, while the llm speech becomes more and more ubiquitous and unhinged.

[Cthulhu_]

> AI has a distinct, low-quality and unnecessarily sensational writing style, which signals low effort post first of all

Yes, but what if it doesn't? Anyone can create a model or prompt to adjust the tone, remove the tells that people are now on the lookout for, etc.

And it's self-reinforcing, because the people that are on the lookout will always consider themselves to be right in their judgment, also because of survivorship bias - they won't know if they missed an AI generated comment/post unless it's revealed.

This is one reason why I'm in favor of tagging AI generated posts, because at one point we can't tell. If we can't tell, does it still matter? I don't know, but I'd like to know anyway.

Of course, it can't be mandated, so if the submitter doesn't tag it and the readers can't tell the difference, it's futile.

[tripzilch]

people could do this, but they don't. and if they did spend that effort, it would be better. but as it is now, the vast majority of AI written text, especially "documentation" of software projects, has this style signifying communication was outsourced to an AI.

we are not at the point where "we can't tell", and if we actually couldn't tell we wouldn't be complaining about it