Over at Camels with Hammers, Daniel demonstrated that the “Argument from Personal Incredulity” is not restricted to mysterians or theists. I tried to post this as a comment, but for some reason it wasn’t approved, so I’ll repeat it here:
I share the suspicion that robots could not have an internal, subjective side of experience.
To which I responded:
Sounds like an argument from personal incredulity to me. Let’s unpack it a bit. Are you a dualist? If so, then you can certainly argue that there is some essential difference between a human brain and a robot brain; you just have the thorny question of explaining how it is causally efficacious. If not, then you have a different kind of hard problem: to explain why it is impossible to create a synthetic substrate which can function in all the functionally relevant ways that the spongy grey stuff in your skull does.
Now perhaps you are arguing that in practice based on currently available technology, such a thing is impossible. Even there, I suspect that you are indulging in a little species-chauvinism. We humans have a natural tendency to exaggerate the competence of our brain function. We actually observe, reason, remember, pattern-match, infer and deduce far less than we imagine we do, with much less accuracy or consistency, and we confabulate like crazy to fill in the gaps when this becomes apparent.
Introspection is simply a form of feedback loop: a process in which a part of a complex system monitors and reasons about the operations of the system itself. Such feedback loops occur all over nature – and engineering. (Check under the hood of your car.) They are fundamental to basic processes like homeostasis, learning, and troubleshooting. It is highly implausible that a sophisticated robot would not incorporate such design patterns.