I don't view it in a black and white dichotomy of trust and distrust. I view it in terms of Occam's razor - simplicity is more probable than complexity - the more complex a conspiracy the less likely it could be successful and therefore less likely to be true. I view social systems as microcosms of human nature which is both virtuous and corrupt. Yet, for the most part people expect goodness, and will rat you out if you are bad, therefore, there is social pressure towards goodness within systems. Ideology, foreign involvement, financial incentive, and social biases can corrupt the morality of a system. So trust for a system would be calculated on how many corruption factors exist to offset the natural human bias for goodness. After a trust baseline is set, I would set my confidence level in that trust calculation to be equivalent to the amount of evidence supporting my calculation. If I have minimal evidence them I can only have minimal faith in my trust bias. The precalculated trust level would have to be verified by evidence of certain actions classifiable into good and bad actions, forming a good actions to total actions ratio. For every ambiguous action, I would need to measure all possible explanations to verify ill intent. If ill intent cannot be verified then that action would need to be included in a numerator for the trust ratio, giving the organization the benefit of the doubt. So I could have a range of values 80% verifiable good actions, 10% benefit of the doubt ambiguous actions, and 10% bad actions. Based on the calculations I would look for the pattern throughout history and see if these ratios are improving or worsening, and look for ways to manage these ratios.
Merely dropping your trust to zero and going the revolutionary anarchist route to me seems immeasurably unwise.