"Prepared to be shocked... I would not be surprised at all.
I've been developing embedded systems for over 30 years"
Same here (different part of the safety critical industry). Back then, things were a lot simpler, a lot less shiny, and in general a lot more trustworthy (at least where I was working, which had its own unique afaik way of doing things). Nowadays, if it's not shiny it can't be good, regardless of how much unnecessary and untrustworthy complexity has been introduced over the years, and how poor the product is functionally in comparison with its decades-old predecessors.
But the industry regulator seems to think this state of things is OK so it must be OK, right? All the process-focused audit-trail boxes are ticked so nothing can go wrong, can it. Les Hatton may disagree but what does he know.
Or do workers with a clue just shut up because they need the money, and there's no place in the company for people who "are not team players".
Here's a 1997 paper from (now Professor) Jonathon Bowen on The Ethics of Safety Critical Systems. Struck me as interesting reading:
The names Feynman (Challenger inquiry) and Haddon-Cave (Nimrod inquiry and several others) also come into this picture somewhere.