It may be that this is closed mindedness, but any answer, from my limited, un-scholarly, un-AI-learned, perspective, to all questions as to wether AI can surpass us is: No.
Just, no. We, superior to AI, cannot create AI that is superior to us. The very name says it all: Artificial. It is inferior to our spirits. We contain possibilities that no machine could ever use, if this machine is independent, not a fusion of human+machine, which would be a way to better explore our potential.
Maybe.. but your 'spirit' or IUOC took a FWAU within this PMR in the *same fashion that another IUOC would take a FWAU within a PMR that we could simulate... And while of course we do not have a Conscious 'Sims' as of yet, it is not far away. See Moore's Law:(Wikipedia): "Moore's law is the observation that over the history of computing hardware, the number of transistors on integrated circuits doubles approximately every four years. The period often quoted as "18 months" is due to Intel executive David House, who predicted that period for a doubling in chip performance (being a combination of the effect of more transistors and their being faster).["
So it's exponential growth, and soon perhaps we will have "conscious" programs within computers, likely quantum computers. I think they would evolve to be as smart or smarter than us. Why not?
The only barrier would be 'creating' the computer that would be capable of "housing" something as complex as consciousness(Tom provides the description for that), and poof, there it is.
Maybe my description of a.i. is outside of the bounds of robots and therefore illegitimate to your point but I think it's maybe the more likely scenario for how we will witness 'intelligence' in our objective creations.