But unlike an automated vehicle, when a black swan event does occur in a human driver you can't disassemble their thought process, correct it, and apply that correction to all other human drivers unfortunately.

yes, but we can hold that person accountable.

are we going to jail some programmer because the missed a corner case and created a fatality.

ultimately what s problematic is what capitalism usually does which is externalize the costs via techno washing.

Maybe not the programmer (unless we get to position where they are licensed like other engineering disciplines), but we can hold a company liable. Some companies have already stated they will accept "full liability" [1]. That reference is nearly a decade old, and some companies have admitted AV is a much harder problem than they initially anticipated, so I'm wondering if they would walk that statement back now.

[1] https://www.motortrend.com/news/volvo-will-accept-full-liabi...

right, but as far as I know, non of the current self driving testbeds are actively insured and liable. maybe it's under reported, but if you used FEMA value of life, you should have seen multiple million dollar payouts.