They will be very impressive mimickers, but they won't be "self-aware".
That may be true and it also may depend on where we consciously lead AI. I guess it also depends on how far into the future we look and if human technology continues to develop. I think Ray Kurzweil's singularity timeframe is ridiculously optimistic but I also think self-aware AI is within the realm of the possible if that's something enough money and time is spent working on. Of course I have no proof of this but rather an educated guess based on my research. At this point your educated guess in the other direction is every bit as valid as mine is. I do think that human and other self aware creatures on earth are simply biological machines and given enough time it can also happen with machines. Evolution is something I've very interested in and something I've invested a fair amount of time studying. I think that evolution will continue to be used to develop more and more complex learning machines until consciousness occurs. I do not however see this as an inevitable outcome.