I'll be moderating a panel at the Telematics Update show in Detroit next week on "The Insurance Piece of the Mobility Puzzle".  For this panel we are looking at "new mobility", both due to the sharing economy (Uber, Lyft, etc) and vehicle automation.  Insurance is often cited as one of the big issues that needs to be solved with automated vehicles.  My oldest son is about to start driving and particular conundrum for me is how to judge his skill.  If he usually drives a highly automated vehicle (soon!), and has no accidents, should I trust him in my old manual shift vehicles?  What's an insurer to do?

On the panel we all agree that the solution to the mobility puzzle is data from the vehicle.  (See the LinkedIn discussion here.)  We have yet to agree on what that data needs to be, on what data needs to come from the vehicle and what data can be obtained from cell phones, or even Apple Play, Android Auto and the like.  My experience is that OEMs are willing in theory to share basic data (location, emissions...), but are not even contemplating sharing the status of ADAS systems and the activation of automatic braking.  

On the other hand, some folks from the CAS automated vehicle group told me an interesting fact.  If Mercedes were to introduce a high end (low volume) vehicle that NEVER had a crash- after 4 years the insurance rate would be only 14% less than the equivalent crashing vehicle.  Seems like there should be a better deal here.