The human brain comes with built-in hardware for dealing with social challenges. It is probably not 100% trustworthy for large-scale society because of the recent appearance of such a structure [1], but it exists.
There is obviously no such built-in hardware support for analyzing software-imposed risks. If such hardware existed, it's not actually the consumer's behavior that would change, it would be the software producers, because the software engineers would feel that bit of their brain poking them and produce much less insecure software in the first place.
So, I don't think the argument translates as well as you say in real life.
[1]: Proof: There exist a large number of people who disagree with $YOU about the optimum nature of government. You are self-evidently correct, of course, and they are wrong. So a lot of people are running on self-evidently faulty hardware for large-scale society.