7 ms·

I am not talking about these people using FSD on and off like it happens now with Tesla. AS it stands that would be a failure, it's just a glorified cruise cont

I am not talking about these people using FSD on and off like it happens now with Tesla. AS it stands that would be a failure, it's just a glorified cruise control I am talking about fully giving control to the car and preventing humans from interacting, after clear statistical evidence of autopilot being better than the avg. driver. Given that by definition the majority of drivers are avg. then society would be better off summoning Skynet and putting it at the wheel. It's a clear cut reasoning that works in theory. Better than avg. doesn't mean that accidents won't happen though. Accidents will still happen and as soon as they start to happen the whole thing will be reversed. That's because people don't consider themselves the "avg. driver" and would rather bet on themselves being better than the avg. driver than the computer being better than that . Even 95 yr old people consider themselves the local Lewis Hamilton. You raised the example of FSD purchases, I cite you the counterexample of vaccines. Vaccines are much much much better than the immune system against COVID, if it were to be close you'd have much less adoption because skepticism is not linear. When in doubt , people bet on themselves or the status quo. In this case the status quo is literally betting on themselves so you have a double effect there too

shkkmo 4y ago
You are shifting your own goalposts here. You were specifically arguing that perfection is the only bar. You have now redefined "perfection" as meaning "above average", but even with that redefinition, your argument as stated still doesn't match reality. People are already letting cars drive for them. They are doing it even when they are not allowed to take over for the car (i.e. Waymo in AZ). Clearly some people have lower bars so your claim that the above average safety is required for anyone to give up control of the vehicle they are riding in to a computer is false. Reading between the lines of what you are saying, is seems that what you are actually trying to argue is more along the lines of "achieving (significantly?) above-average safety for self-driving systems is crucial to achieving mass adoption of self-driving technology." That's a much hard to disprove statement (and one I suspect it is at least partly true given varying levels of "cruciality").

shkkmo 4y ago

You are shifting your own goalposts here. You were specifically arguing that perfection is the only bar. You have now redefined "perfection" as meaning "above average", but even with that redefinition, your argument as stated still doesn't match reality. People are already letting cars drive for them. They are doing it even when they are not allowed to take over for the car (i.e. Waymo in AZ). Clearly some people have lower bars so your claim that the above average safety is required for anyone to give up control of the vehicle they are riding in to a computer is false. Reading between the lines of what you are saying, is seems that what you are actually trying to argue is more along the lines of "achieving (significantly?) above-average safety for self-driving systems is crucial to achieving mass adoption of self-driving technology." That's a much hard to disprove statement (and one I suspect it is at least partly true given varying levels of "cruciality").