I've also noticed that very obviously LLM-generated comments are called out, and the community tends to agree, but those that have any plausible deniability are given far too much leniency, and people will over-index on the guidelines to give them the benefit of the doubt.
I don't think a captcha is the solution, as it'll degrade conversation by an OOM though.
Normally I'd agree, but we have shadowbans, which really irks me.
Almost everyone banned on HN is banned publicly, with a public message explaining why.
I would love for this to be the case, however I quite extensively investigate this phenomenon and this does not match what I've seen. I'd like for us to be better than shadowbans. In some cases, I don't even get to vouch, it's just a comment that is banned-banned. It feels the worst when they're saying something substantive to the conversation and we have no means to surface the comment.
Some type of annual amnesty consideration or something of that nature is in order, or soon we'll recreate other echo chambers that are slowly fading out.