Depending on how the driverless vehicles are programmed to respond to a person (if they give any sort of feedback or not), I might feel confused in situations where I was trying to communicate (how to wave-along a computer, or how to nod to acknowledge the computer is yielding to you?)—Otherwise yes, I would feel comfortable.
A car doesn’t get tired, or irritable, or aggressive, or distracted, or scared, etc… in that way, they are much more predictable than people.
And we’ve gotten computers pretty good at recognizing what they are seeing, so I feel pretty confident that a driverless car would keep up with any changes in the environment.
I imagine, too, that the durability would be fine. Modern cars today already have computer chips, sensors, etc., and those hold up well—some of those are cameras detecting motion in blind spots, or sensors on wheels detecting when road conditions change—so I assume any self-driving components would hold up comparably well?
For sure I trust a car to park better than me (I’m an awful parker), and I’d trust the driverless car at least as much as myself when I’m alert, definitely more so when I’m having an off day.
Going to the VCR analogy… even if people never spent the time to figure out how to set the clock, I imagine they knew how to insert VHS tapes and get them playing. It seems like it would be a similar situation with the car… people would know at least the basics. They’d know how to start the car, how to input their directions, the general requirements for maintenance, etc… maybe they never set the clock on the dash, but then, I can never remember how to set the clock on my car radio right now; I always have to pull out the directions twice a year, for spring forward and fall back. :P
Also, here’s is an entertaining read on what it’s like to ride in a google car (by the oatmeal)