Want I’d want to see is a focus on level 5 autonomous driving from day 1. (Edit: Even if the system is level 4 to start with.) Yes the current coverage area is limited, but if you live in one of those cities the coverage area is easily large enough to be useful.
Oddly enough I think this is one of the few times when a subscription model makes sense. The current approach has a fallback call center which can give the cars driving directions in unexpectedly situations, which could be supported by either a monthly subscription or low hourly fee. Similarly move out of the coverage area and stop paying etc.
As someone who's blind I've made this argument in the passed. I don't need 100% success as long as the failure mode won't injure me. Seven years ago I would have loved a car that would have driven me to and from work 95% of the time, and refused to take the other 5% if the weather forecast was bad enough that the self driving wouldn't work correctly. I'd also be fine with the car pulling over to the side of the road if it got confused and waiting for someone remote to take control and drive until it was out of the situation where autonomous driving wouldn't work. Given the fact that I now work remote and am married to someone who drives if you told me I could by a car with autonomous driving for $50000 now I don't think I'd do it. I'm interested to see just how good autonomous driving gets and if it drives down the prices of taxi services. At this point I'd rather see an autonomous taxi service offering lower rates then Uber instead of buying my own car with autonomous driving.
I think you may mean Level 4. The difference between 4 and 5 is that 5 doesn't have any territory/environmental constraints, but you said you don't mind those.
If they require high speed cellular service then the system can’t scale to level 5 driving. Add a Starlink dish on top and the hardware could eventually scale to the entire continental US etc.
And I actually agree about the subscription model too. It's one of the rare cases where it feels practical: pay while you're in the coverage zone, pause when you're not
Oddly enough I think this is one of the few times when a subscription model makes sense. The current approach has a fallback call center which can give the cars driving directions in unexpectedly situations, which could be supported by either a monthly subscription or low hourly fee. Similarly move out of the coverage area and stop paying etc.