Humans fail chaotically. AVs, at least Waymos, fail safely. In urban environments, there is a case to be made within a few years for banning (or making it massively cumbersome) to manually pilot a private vehicle.
People will trust them by familiarity, not by any objective measures. Their trust grow as they see them more often, without bad accidents.
Would you rather have a X% chance of a loved one dying from a human-category mistake? Or a 0.5X% chance of a loved one dying from a ML-category mistake? I would take the latter in a heartbeat. Why waste time optimizing for "categories" when you can optimize for your loved ones not dying
You and I are thinking with a purely logical thought process, but most people don’t reason that way.
A lot of people want to be able to hold someone accountable, and that is more important than the pure probabilities.
They want to be able to blame a person for making a mistake.
I live East of DTLA and twice I’ve seen people driving them around, presumably to map those areas out. It’s pretty annoying that I can’t use them for my late night commutes yet, but I’m eagerly looking forward to when I can!
So... not so sure. Every AV I've ever been in has driven incredibly cautiously compared to a human driver.