>Because your neighbor's problem eventually becomes your problem.
This is an argument for doing something, not necessarily for putting a burden on individuals to counter extremist and murderous ideology - especially those who are most likely to be targeted by it. It may be a moral imperative, but moral imperatives really are sensetive to a variety of individual circumstances to exercise, because they only work by way of moral obligations and moral responsibilities - both of which are only parts of rational decision making.
Someone who has had a strong past of experiencing racism may balk at the idea of starting a dialogue with an extremist whose entire ideology is founded on the view that he is not even worthy of consideration as a civilized person.
I firmly believe that most people lack the capacity of mentorship, or if they have it, then they are not terribly experienced with it. The fact that professional mentors take so much time and effort to nurture someone who they know is already receptive and still fail should be a testament to a significantly worse ability in the common person to do that, especially when, as I mentioned, communication is a two way street. What makes you think the mentors will be any less susceptible to radicalization than the people they are mentoring?
The biggest problem is that most people do not have the requisite knowledge to be such mentors. They may know racism is wrong, but maybe they can't explain it sufficiently well. They may appeal to humanity, but it may not get far to do so with someone whose ideology for the past two decades has been the fundamental inhumanity of the person they're talking to. Needless to say, all this completely leaves out the fact that I really don't think most extremists are keeping an open mind in the first place.
>When you start editorializing an individual's speech rather than debating them, it's a slippery-slope down the road to censorship and fascism.
Slippery slopes need to be justified, not merely supposed. The mechanism needs to be explained and the risk made clear. For example, the mechanism that one form of speech (like hate speech, defined stringently) can be banned. It is not at all obvious that this leads to other ideas being banned; the legislature, for instance, need not and generally does not work on the basis of precedent. Superior courts can override precedent. Similarly, the fact that a hate speech law can be abused is not a good argument; plenty of laws can be abused, including, say, anti-terrorism laws. This does not mean that anti-terrorism laws should be done away with.
>Having venues that can debate and debunk bad ideas with better better ones is preferable to preaching to the choir.
This assumes the debating and debunking works and wins out over rhetoric. We're not dealing with pure ideas, we're dealing with two forms of expression of those ideas: firstly, hate speech which is a generally non-argumentative and derogatory expression of the idea designed to be hurtful; secondly, speeches, texts, videos, music, etc. which combines rhetoric and argumentation. Debate does nothing for the first category. It may do something for the latter category, but it's unclear if it does so at the scale or effectiveness to satisfy the moral obligation you pointed out.
>Someone can't influence or the change minds of people they refuse to talk or listen to.
The point of the laws discussed is not to change peoples' minds, it's to hobble the spread of their ideas.