I was very surprised to learn that according to Less Wrong, the concept of "Zombies" is accepted by a lot of researchers of consciousness: http://lesswrong.com/lw/p7/zombies_zombies/ A zombie here is defined as a person that is exactly like a conscious person, down to every atom, but is not conscious. I think that is complete nonsense.
If I attach a light sensor to my Arduino and then upload code to the arduino that says "if sensor active, flash light", I'd say the arduino has a "qualia" of light. Perceiving light flips some bit in the arduino. That's all there is to it.
I think this just boils down to the old and tired argument that "computers can only do what humans told them to do, therefore they can't be conscious", which is just misinformed. If I tell the computer car to turn left every time it sees a duck, I will have no idea where it will end up.
Also building a "drug me" button into an Arduino seems trivial. You could XOR all your data if the button is pressed, or do whatever else you fancy (even trigger a completely different program). I don't find drug experiences puzzling at all. If you want, overclock the Arduino so that it produces some computing errors. Hey presto, your Arduino has a drug experience.
But if that's the case, why does the first-person point of view exist? Yes, there will be a bunch of people, all supposedly with their first-person points of view, but why are you inside yourself? What is different about your illusion of consciousness such that you are inside it from the first-person point of view?
If consciousness is just an illusion created by chemical reactions, it shouldn't exist at all. Consciousness is a first-person thing, and were it just a bunch of reactions, no one would be able to call themselves I.
But OK, that is just an opinion, I lack the energy to write more atm.