Not the GP, but I'm thinking you could have a system where instead of centralized control, as a user you can subscribe to blocklists provided by the company as

Not the GP, but I'm thinking you could have a system where instead of centralized control, as a user you can subscribe to blocklists provided by the company as well as community-made ones. These would be tasked with the regulation of legal speech, and providing a framework for these lists would be a requirement to be protected under Section 230 of the Communications Decency Act if you meet certain requirements; while illegal speech can still be controlled under traditional centralized moderation. Not sure how legislation would define the requirements for something like this. What would metrics would signal that a platform to have to comply with the regulation? (MAUs?, revenue?, surely having moderation, but should certain kinds of moderation like removing botspam count?) Should a provider be able to silently set-up default lists for new users or should they be prompted? Should the company be required to break down blocklists among different topics (spam/pornography/depictions of violence/a goose with a campfire in the background) and how should the boundaries of topics be defined if so? What kind of functionality should be considered the bare minimum for a framework to build community-curated blocklists? Mind you, I still am not sure regulation is the way to go. But the truth is that these platforms wield a scary amount of power over the psyche of people to extents that would've made the Stasi drool all over the place, and it does have the ability of corrupting democracy far more than biased TV channels and newspapers since it's no longer Fox News or CNN presenting you something, it's your own acquaintances that get some of their voices artificially boosted while some others are silenced; literally distorting a huge amount of voters' own grasp on reality.