You used to have what was essentially airgapped and self contained.
But then feature x needed an ongoing net connection, and it happens to run on the same soc as feature y that talk to the can bus, and boom.
Neither of the teams responsible for the features considers that something can jump from x to y, almost like an illness jumps between species.
Damn it, the other day HN linked to an article on how VMs sharing hardware could talk to each other using the CPU cache.