As far as I understand, censorship like this is primarily done by "incentivizing" internet platforms to patrol their own content. Some of this is done by algori

As far as I understand, censorship like this is primarily done by "incentivizing" internet platforms to patrol their own content. Some of this is done by algorithm, but a lot is done by human moderators who are very low on the command chain but have to exercise personal judgement on what to ban and what not to ban. Many harmless words have already been banned in various platforms due to mis-classification errors, and if underground activism picks up pace, maybe it'll eventually crack the rather crude censorship method at scale.