> the interesting numbers would be fatalities per driven mile without autopilot engaged vs miles with autopilot engaged Exactly. I imagine hell freezes over be

> the interesting numbers would be fatalities per driven mile without autopilot engaged vs miles with autopilot engaged Exactly. I imagine hell freezes over before we see such data. Tesla did not even provide complete data to NHTSA [1] [1] https://www.thedrive.com/tech/26455/nhtsas-flawed-autopilot-safety-study-unmasked https://www.thedrive.com/tech/26455/nhtsas-flawed-autopilot-...