I had two accounts banned from BlueSky and they didn't say why. One was parodying Donald Trump so fair enough if they don't want content like that, and they told me it was banned for impersonating Donald Trump. The other, no idea at all because I don't think I even tweeted anything very controversial, and the email was just a very generic "you violated terms of service". My third account was not banned, but I don't use BlueSky any more. It's not a ban-evasion ban, since they're logged in together in the same web browser, with the menu to switch accounts active, and yet my third account was not banned.
My point of sharing this info is that BlueSky is not a user-driven moderation system. It arbitrarily and centrally bans accounts, just like Twitter.
You're right, Bluesky moderation is centralized. Unless content is served p2p, some moderation has to be centralized. At the end of the day, there's a server serving content and that server operator is legally obligated to remove illegal material.
Hopefully, atproto + community will provide alternatives for moderation services. Work is being done on this, we'll see what we end up getting.
I feel that a competitive ecosystem of moderation services is probably the best answer we can hope for to that inherently messy problem.
My point of sharing this info is that BlueSky is not a user-driven moderation system. It arbitrarily and centrally bans accounts, just like Twitter.