What makes you think that? Given that we have one class of systems that have evolved general intelligence I can see no reason why we can't engineer others. This may, of course, be rather difficult but I'm not aware of any good argument that engineering a general intelligence is impossible for fundamental physical or logical reasons.
Well, it's based on a bit of a philosophical cop-out, but basically anything that claims to be strong AI will always only be a philosophical zombie. It might behave intelligent, it might claim to be intelligent and conscious, but it's impossible to prove that it actually is intelligent and conscious, since it's impossible to objectively measure or describe consciousness.
In short, I'm with Searle. We can never have strong AI, but we can certainly have weak AI.
What makes you think that? Given that we have one class of systems that have evolved general intelligence I can see no reason why we can't engineer others. This may, of course, be rather difficult but I'm not aware of any good argument that engineering a general intelligence is impossible for fundamental physical or logical reasons.