A 5 year old post I made on FB organizing support for a local synagogue that got tagged with antisemitic graffiti was just flagged for hate speech. (Yes, really.) Is the algorithm wrecked to the point where it can't tell the diff between hate speech+organizing against hate speech? Or was it a troll?
Follow
@GWillow I've seen and heard trustworthy secondhand reports of similar situations as far back as one decade ago (the one i most vividly remember was a post about police brutality that mentioned threats the police made that was removed "because it contained threats addressed at another person"). I suspected that this is a result of some very surface-level pattern matching by underpaid moderators triggered by a report from a troll.
