A few years old now, but as I contemplate fediverse growth, this essay by @mmasnick is a thinky read for those in the instance moderation or community admin space.
Our hands-on, know-your-clients moderation paradigm can't survive 6- or 7-figure user numbers. But algorithmic moderation, as is presumed to be at least the first pass methodology in this essay, is bound to have both false-positive and false-negative results at levels which are in absolute terms, horrible