That's not what I'm suggesting.

The problem is ultimately anonymity, and people trusting anonymous posts as face value.

If we had a rigorous identity layer, a lot of this would be resolved.

People also trust non-anonymous posts. You just have to form your own view based on several sources, and try to verify / dismiss some of them. Put everything together, hopefully not like in the movies (wall full of pictures, thumbtacks, and rope)

Sure identified sources can be bad actors too, but that doesn't really matter w.r.t free speech.

For one, the effects are asymmetric. Anonymous bad actors can do more harm than identified sources can do good. Identity discourages bad behavior.

Second, free speech and other ideals of liberalism assume that people are fundamentally good. So if the identified sources are openly bad actors such that they out weigh the good actors, then you're kind of screwed regardless.