From my own experience it mostly works but needs to be smarter. They need an automated way to prove ownership of content or an account on a third party site to stop its nagging and eventual flagging. I've seen cases where people were as desperate as Michelle to stop being bugged and had concrete evidence that they were who they said they are. Something clearly wrong there.
I think all bots should have owners with proof of brain that can be challenged. I think all bots should issue warnings before taking punitive action. I think all bots should have something at stake so they can be reigned in.
I don't know how you enforce these things but those are my thoughts. How about you? Are you a cleaner fan?
Me? the only time i approach Steemcleaner is out of frustration, you know.. those days when you see the same damn account shitposting for the straight 50th times. So it does have it's uses.
But yeah i understand your point of view, its a bot after all, not an AI. so sometimes it follows the set of rules so closely that all subjectivity is lost and innocent people do get in the crossfire.
It's something that just needs ironing out i guess.
My opinion is that if everyone has the awareness to properly use the downvote button under the correct circumstances, Steemit would have been a way better place.
There's just seems to have this air of stigma towards the red flag. Which i dont think is fair. There is nothing wrong with expressing disagreement of rewards or better yet, calling out reprehensible acts. Do note that when i say "disagreement of rewards" its entirely subjective to what the voter thinks as worthy, same as upvotes.
I do think the downvote bots we see now are not the "final" version and they do serve a greater purpose towards encouraging people to better police spam and unethical gaming of the rewards pool. Ideally in future we don't need it at all.