The cause(s) is/are actually disputed. It is likely that there were multiple causes. The specific bug I am referring to is this stack overflow bug:
Cars will always have issues with physical bugs (think exploding airbags recently) which cannot be fixed by a magical sw patch. Therefore, real recalls must be effective and timely.
The fact that some hardware bugs can't be fixed by OTA updates doesn't mean that we should fix the ones we can through OTA.
Yes, some problems will require people to bring cars into dealerships. And some people will no doubt experience those problems within the time frame of after knowing about the problem and before the car was brought into the dealership to be fixed.
The recalls can be effective and timely relative to the standard of a recall. They can't be timely compared to an OTA update.
I might argue that making OTA the norm will mean recalls requiring people to come in for physical objects to be replaced will get sloppier.
That seems highly speculative. I could just as easily argue that having less recalls wil allow the workers to concentrate their time and energy on performing the recalls that still happen properly. This is something that needs to be decided by empirical data and statistics, not speculation.
And I maintain by allowing OTA access to the machine means there will be bad actors who can hack the machine. Nothing is unhackable.
It doesn't need to be "unhackable" (which you could never prove anyway). It needs to save more lives than it costs to be worthwhile.
There are already so many things that are capable of wireless communication that our lives depend on. Why not focus on making the security better and better, rather than avoiding it.
Even commercial aircraft with hundreds of people on board are not unhackable. Should we strip all the communication equipment out of commercial aircraft because of the risk of being hacked?