Deep learning based systems are at least trustable as working animals I think... They share a lot of similarities. You can not know how it thinks, but you can train them, observe them, and then trust them to a certain level.
I am quite interested in the difference between people's intuition about things. The more complex a system is, the less I tend to think about it being designed.. It is just too beautiful to be a crafted artifact.
This is rather a math-specific operation (type conversion on math operators), not the scope of general purpose language. A lot of dependent typed language can achieve this naturally and most languages with ADT can simulate this.
A commercial project aims for profit, not public service. The previous post explains why they are popular demands. And it is natural for commercial project to satisfy popular demands.