I think handling of spam or ads will require "micro task queues", where individuals with higher trust can work (= earn sats) to resolve those. For example if I see spam I could report it. That immediately puts that post in the "spam task queue", it would cost me 10 sats to do that. Then someone with higher trust can work on the spam task queue and decide whether the post is a spam or not. This individual would be rewarded e.g. 5 sats. (and maybe original reporter could also be rewarded some sats back for correct reporting). The spam task queue could be prioritized by the amount of sats that people pay to mark it as spam.

When the post is confirmed as spam it could be then marked as such in the UI and users can decide whether they would like to include "spam" posts or not in their view. E.g. this is not censoring, but rather letting the user to decide whether to filter some type of post out or not...

The beauty of this is that the sats flow with the incentives - common users pay for seeing less spam, other users get paid for doing spam resolution, etc...