This comment should have 99 upvotes and should be moved to "Main" as a separate article. Then we should link it whenever the same topic appears again.

Reversing group-think is like reversing stupidity, or like an underconfidence at group level. It can be done. It can be interesting. But I prefer reading rational people's best estimates of reality. And I prefer disagreement based on genuine experience and belief, not because someone has felt a duty to artificially maintain diversity.

If you disagree with whatever, for example many-worlds interpretation, say it. Say "I disagree because of X and Y". Or say "I disagree, because if feels wrong, and because many people disagree, including some experts in the field (which is a good Bayesian evidence)". That's all OK. But don't say or imply things like "we should attract more people who disagree with many-world interpretation, to keep our discussion balanced". That is manipulating evidence.

If anything, we should discuss wider range of topics. Then naturally we will attract people who agree with N-1 topics, and disagree with 1 topic; and they will say it, and we will know they mean it.