So a key fact from the comments after the paper was published I expect all the blogs to miss: I asked the author how many of the reported CSAM hits were blocked

So a key fact from the comments after the paper was published I expect all the blogs to miss: I asked the author how many of the reported CSAM hits were blocked by servers that include a very common "bare minimum" blocklist for bad servers, and he checked and said 87%. So the majority of the CSAM that they detected never reaches most of the fediverse. Furthermore, the methodology of the study relies on sorta Mastodon's firehose, which is instantaneous. That means the hits they catch are before moderators get a chance to handle them. In most of the remaining percentage, they likely get reported and blocked quickly. So most of the fediverse is blocking the vast majority of this garbage already. Sure, there's still a lot of work to do, and plenty of ideas how to improve moderation, but the fediverse doesn't have VCs. Donate to Mastodon if you want to help fund work on this stuff.