Interesting approach to moderating bad behavior in online games:

[Garlyle]

I take it you've not actually played League of Legends and seen the level of bad behavior/bad sportsmanship ingame?
The complaints send to the ai are user made reports, which are actually abused a lot.

I play a lot of League, actually. The argument that it's cleaned up significantly isn't made off of guesswork; it's both from data that's been released and personal experience as well.

The report management system tracks things like how frequently you're using reports, your own attitude in the games you used reports in, whether the players you're reporting are reported by others, the accuracy of the reports you've made in the past, etc - and can put extremely low weight or even outright ignore reports by players who are trying to weaponize it. For all the fears people have about an automated system, the amount of times I've seen actual wrongful bannings that had to be overturned can still be counted on one hand - which is impressive for a game that's been able to claim more players than World of Warcraft at one point.

I've been tracking player behaviour related studies a lot in recent years because it's utterly fascinating, and League's one of the biggest contenders in trying to find ways to combat it instead of just letting it be.

It's still there.

Still there? Yes. Negative behaviour is, sadly, never going to disappear from online gaming, or at least not for a long time, because people still assume it's the accepted standard (it's actually not, and there's a lot to be said about why we think it is). And if the only measure of success it "it no longer ever happens", then you're expecting the impossible.
The actual question is, have these efforts and systems made a significant impact on things? Absolutely. Can they make an impact elsewhere? Absolutely.