Hmm, not quite what I thought it would be. It's just a javascript widget that let's users report content to your own team.

I know that Facebook (and I assume the other big players) run teams (from what I read not excellently paid and without enough psychological support ) that actually review/filter objectionable content and I am guessing they're throwing their automatic detection systems at it as well.

Anyone aware of a company offering the actual filtering system AAS ?

Not a company, but I run https://blogspam.net/ which is a real-time API for testing comment-submissions for spam.

It uses a bunch of (naive) rules to classify content, and also allows (re)training.