"Chemically driven", amounts to something like - the presence of one chemical, causes another chemical to be released, which stimulates electro-chemical impulses in nerves, which sends a signal to the brain.
You could have something analogous to the initial chemical chain, or you could just build something that would receive the equivilent of the electro-chemical signals (probably in electrical or photonic form rather than as electro chemical signals), or you could skip that and just create something that would respond as if the signals have been received.
Simulating hunger (not just the appearance of hunger, or the reactions that an organism has to hunger, but the actual feeling of hunger), should be easier than creating sentient "true AI". If it really is necessary for "true AI"/artificial sentience, than I don't see why it can't be done as well.
I can see why it, and more generally "true AI" is very difficult, I can see the argument that none of our current attempts are even heading in the right direction. It would not shock me if I don't live to see it (although it also would not shock me if I do live to see it), but I very much doubt that in the very very long run, hunger is going to be such an obstacle. |
