The problem is that skepticism on its own isn't a route to the truth. An embarrassing number of people "doing their own research" are doing on reddit, not e.g. arxiv.
Just as well-founded loop variants help avoid nontermination, making strictly closer-to-ground double checks helps avoid echo chambers.
(and frankly, "do your own research", to me, is a clear sibboleth translating to "enter our echo chamber")