The way of vulgar skepticism is to assume something doesn't exist until it has been proven to exist.
In cases like the Lock Ness monster or invisible pink dragons in my garage, that approach happens to yield good results (and so would blanket generic cynicism.) But it falls apart quick when you stray very far from the fantastical, e.g. "masks have not yet been proven to work, therefore masks don't work."
A bayesian approach to skepticism is much better than vulgar skepticism. Bayes asks the skeptic to consider the general outlandishness of a possibility P(A), the basic likelihood of apparent evidence occurring regardless of the hypothesis P(B), and the likelihood of there being evidence at all even if the hypothesis were correct P(B|A).