It's not about what "people don't want to see".
(Way too) many people want to see CSAM and yet we ban it simply because of how corrosive it is to society. And that's true for a ton of the other stuff. The lynch mobs that have taken place on Whatsapp for example - you can't just put a feature like global groups or unlimited forwarding of messages on the Internet without taking into account that people can and will abuse it, and then not even have some sort of report system where people can say "hey. have a look at this". That's basic ethics review that should have caught this abuse potential and suggested mitigations before people lost their lives.
And the lack of moderation around hate speech of all kind is similarly corrosive. Sure, in the US it might be legal to go on the street and call for a "peaceful ethnic cleansing". But if you do so, you will likely eventually get insta-sanctioned by someone just like Richard Spencer [1]. On the Internet however? Nothing, other than maybe getting a shitstorm. There is no feedback mechanism in the libertarian/ultra-free-speech internet to signal to people that what they are peddling is not the mainstream view, no way of triggering some self reflection.