Taking it as a fact that AI detectors are 100% effective and accurate is just as crazy as saying everything an AI says is 100% valid information. When I look at people who rely on AI detectors I see an extension of someone who also relies on an AI.
The fact is these AI detectors follow a regular pattern with the aim to burst an AI generated text and it's not 100% accurate. I see it more like a speculation not a factual feedback.
I disagree, more often than not, these detectors have successfully located numerous spammers, I don't recall ever seeing a case in the AI Report Thread where someone would successfully defend themselves and prove that they weren't using one. The majority of cases are from newbies spamming copy pasted gibberish that can be distinguished that it's AI just by reading it. A few older members who were caught with AI either quit without any explanation or nothing that could remotely backup that they weren't using one.
This is the reason why we're using multiple detectors over a variety of posts, we're not going to accuse someone just by one post.