I don't disagree they should probably be more proactive, but it's tricky because "X is a horrible person" can be legitimate articles too; banning domains is a whack-a-mole, and of course you'll get complaints about "censorship".
Filtering against slander seems like reasonably simple and not one that different from what they already. They're already doing whack-a-mole against a whole variety of things.
But yeah, shitty websites suing their ways onto Google search results with "free speech" complaints may one thing that is making this hard but I don't know enough about that situation to say with any certainty.
It's not.
> not one that different from what they already.
True, and they're pretty bad at all those other things too.
Unfortunately the societal costs of false-positives and false-negatives are not the same as Google's cost of false-positives and false-negatives.