I'm curious why it is assumed by so many that it is certain that a sufficiently intelligent machine could produce even better machines. This is certainly at le

I'm curious why it is assumed by so many that it is certain that a sufficiently intelligent machine could produce even better machines. This is certainly at least not demonstrated, and all evidence I've seen for it is extremely weak.