Absolutely. Our heuristics for judging human output are useless with LLMs. We can either trust it blindly, or tediously pick over every word (guess which one people do). I've watched this cause havoc over and over at my job (I work with many different teams, one at a time).

AI signatures don't mean low quality, they just mean AI. And humans do use them (I have always used the common AI signatures). And yes, humans produce good-looking garbage, but much more commonly they produce bad-looking garbage. This is all tangential to the point.

Yes. Without a doubt. An example from the software world: I came across a Rust rewrite of scikit-learn the other day that looked impressive at first glance but was full of correctness bugs upon further inspection. In the past, this kind of shoddy work would have had a code smell that was easy to clock. But now, thanks to LLMs, these kinds of projects appear to be professionally done when in reality it's just a beautiful facade in front of a pile of shit.