SageBrush said:
You have NO IDEA what the risk is, because GM has not figured out the cause(s)
I would just like to point out that you don't need to know the cause to estimate the risk.
I haven't been following so closely, but there have been, what, 5 incidents so far? So one can model the events in a few different ways (constant chance per car per day of its life; increasing chance as the car ages; etc) and for the models which are consistent with the incident pattern so far, get an estimate of the chance of occurrence.
It's not like your car has a 1 in 1000 risk of catching on fire tomorrow. For a basic upper bound, if there were 5 events in the last year, and there are 70,000 cars, that's more like a 1 in 10,000 risk of happening over a year. Which is too high, but in my opinion not so high as to require behavior changes beyond those recommended.
Cheers, Wayne