ppragman
No pasa nada.
I don't know what you mean by this? Because you have to get the stats on how often the system fails, etc.so if the system doesnt work unless you have 100% participation what are the real odds?
What we know, right now is that 1/95 people get killed from car accidents in their life. That's a fact. What we don't know (and I think this is a problem, to be clear) is how often mechanical and software errors could have been saved by a person but resulted in an accident anyway in self-driving vehicles. That's kind of a hard number to reason about... but I'd be willing to bet that the 1/95 stat vastly over powers the likelihood of a human driver saving the day.