You seem to be claiming that the only people who can be trusted in mission critical scenarios are those who have seriously failed with a non-negligible impact before, as though that is a rite of passage to true understanding. How are you defining failure?
I disagree with the idea that meticulous caution can only be learned by personally failing first. I believe it is fully possible to learn and emotionally internalize the necessity of proper protocols, rules and behaviors in a variety of safety and compliance contexts by learning from the failures of predecessors, and without needing to commit those failures personally.
If this were not the case, wouldn't e.g. insurance companies weight risk analysis algorithms more favorably towards drivers with an accident history, and penalize drivers with no accident history? (This is a simplistic example, but you hopefully get the idea). Do pilots need to experience catastrophic crashes in order to achieve true risk awareness? Do companies need to experience serious security breaches in order to take security seriously? (Many of the most secure companies in the world, perhaps most of them, have never had a headlining security vulnerability).
To continue with your analogy, I see no a priori reason to assume that a doctor is incapable of following e.g. comprehensive checklists to limit surgical error without first experiencing a failure that causes a patient's death. Likewise, I see no a priori reason to assume the same doctor would improve by virtue of having a patient die in surgery.
I think that claim requires evidence to be taken seriously, to say nothing of your secondary claim that people who have not failed have "too much bravado" to be trusted with responsibility.