[undergrowth54]

This is a _much_ harder epistemology problem than you are giving it credit for.

Do you know of a way to ensure that a newly-developed effective vaccine has a non-figuratively literally 0% chance that it could cause anti-dependent enhancement or other harms. If so, I'd love to also hear your technique for writing software with absolutely 0 bugs.

Otherwise, we are dependent on trusting institutions to

1. Gather data about the statistical prevalence of harms.

2. Analyse that data and confer with their peers.

3. Broadcast judicious conclusions to the general population.