You can simply blur the post like Quora does and by clicking on it it would reveal. There is already one mechanism in place by voting threshold, but very popular posts in certain subreddits might not be ad-friendly.

I am looking at social media as a form of reinforcement learning, especially the exploration phase; by censoring it the exploration phase is significantly weakened, leading to worse results in the end. So maybe evolutionary argument would be to allow all kinds of crazy ideas (with some requiring a bit more work to read), and let people themselves to decide which ones would be used in exploitation phase? In the past when a single stupid idea was forced upon everyone by authority (e.g. nazis), people didn't have any means to counter it. With open discussion spaces some people might find excellent counterexamples immediately, clearing the issue for millions right away why a certain idea is a really bad one.