Yeah... so I honestly consider this to be a separate (also horrible, for avoidance of doubt) problem than "moderation"? Like, Apple and Google aren't exactly ha

Yeah... so I honestly consider this to be a separate (also horrible, for avoidance of doubt) problem than "moderation"? Like, Apple and Google aren't exactly having a moderation problem when it comes to attempts to curtail CSAM stored on their photos platforms (which has led to their automated flagging systems and then need to scale to the entire world of users appeals and escalation, and then the subsequent concern about "are they going to call the police on me when I fail my appeal?!")... it is more of an attempt to deal with awkward regulatory "think of the children" overreach and hostile American law enforcement. Facebook seems to me to be doing the same thing here and failing. Imagine a world where social networks were built by people who simply didn't care about "engagement" at all and weren't being motivated by ad revenue... I think you could design an end-to-end encrypted version of the system where no-one except your friends--and certainly not the network operators--even knew what you were posting in the first place and they would hopefully be able to avoid installing client-side filters for CSAM (but, with stuff like FOSTA and SESTA, maybe not?). This model should even work, I'd think, for Twitter/Instagram-like broadcast models (though the legal implications of the well-known "secret" key being published and accessible to the network might lead to various problems; you might have to go fully-decentralized).