> This is a different category of freedom. Failures of air safety in manufacture, maintenance, and operation can and do kill innocent parties on the ground in c

> This is a different category of freedom. Failures of air safety in manufacture, maintenance, and operation can and do kill innocent parties on the ground in crashes and in the air in collisions. The vast, vast majority of airline deaths are from people who are employees or paying passengers. "This could conceivably hurt a third party" is, when unquantified, a blank check to regulate anything. All activities have non-zero risk to third parties. > JSX's model was 10 times more likely to be in an accident and 18 times more likely to be fatal than major airlines. This was in line or better than any other year I looked at back to 2008, which is the earliest data the NTSB had. You're missing the part where you actually quantify the benefit. That involves taking the accident rate, computing an expected chance of death per passenger per flight (or mile or whatever), converting to dollars using the statistical value of life (~$10M in the US), and comparing that number to the ticket price. You'll find that for the major airlines it's <$1 per flight. Just ask: how would you know when it was no longer worth it to spend more to drive the risk lower? Any argument of the form "this activity is 10x more dangerous than another activity" is incomplete if you don't know what the appropriate level of danger is for the other activity.