I think of this when I see disagreement about how far autonomous cars are. One the one hand, I can agree that we aren't quite there yet. There might even be som

I think of this when I see disagreement about how far autonomous cars are. One the one hand, I can agree that we aren't quite there yet. There might even be some unknown problems that might add decades to the goal being achieved. On the other hand, there is improvement happening in all the subcomponents. Processing power, hardware, software etc.This is what really makes it hard to predict. Especially if you're not an insider and rely on information mostly from public announcements and news. For example: I hear autopilot 2.0 is not doing so good from some people. While others says it's feature locked on purpose while things are being tested out. Maybe both views has some truth to it, but I can't gage just how good that software actually is.