Research on software reliability is plentiful and throughful, and there are existing tools to use such knowledge in a way to ensure safety of critical systems,

Research on software reliability is plentiful and throughful, and there are existing tools to use such knowledge in a way to ensure safety of critical systems, from development to deploy. The thing is that they most likely ran their calculations and concluded that it's more profitable to cheap out on software development and QA, and pay for insurance or fines, should anything go wrong later. Engineers who devoted their time and effort to learn aforementioned techniques are very unlikely to work for 9$/h, it's more profitable and less stressful to go wash cars or whatever.