|I think you missed the point I was trying to make. Nearly everyone has different skill levels and abilities. Someone who might have good skills with their hands or have excellent eye/hand coordination will be suited for different tasks than someone of lesser aptitude for these tasks. AI should, in theory at least, be able to quickly analyze or test an individual to gauge not only what (and where) they are able to be more successful but also where they may be likely to have more satisfaction and happiness in their work. Someone who might not be the sharpest knife in the drawer may be able to pilot a drone more efficiently or cook better and so on. They may be a better gardener or work better with animals.|
So taking your point about a "smart" person but one who has never even drilled a screw into wood, where do we start? Was it the lack of education, experience, initiative or what? Then jumping into "visualizing software architecture", which is likely going to be done by the AI systems themselves and not by even the smartest individuals, strikes me as way off base.
Some of the "winner take all" concepts may start to be modified as well unless we are going to continue to repeat many past failures.