While I'm a huge lover of Asimov books, I don't think most of the novels represent a probable evolution of how robots would evolve since the rules in the book r

While I'm a huge lover of Asimov books, I don't think most of the novels represent a probable evolution of how robots would evolve since the rules in the book represent distinct rules based upon something that'd be fuzzy at best to evaluate, and even if that was possible we'd be more interested in debugging those failures rather than having a brain fry itself. That said, once the capabilities to add reasoning and concretization of fuzzy concepts is within the realm of possibilities maybe it's a sane approach to install those rules, still the present trajectory doesn't really imply it would go there.