[jbotz]

Actually I do; but... 1) such a Turing machine may have to be a lot more powerful than anything we currently have. With or without invoking QM mechanisms, there is reason to believe that every single neuron does a lot more computation than our simplistic models in current ML neural nets. 2) it may not be possible to "program" a machine to be conscious in the way we feel consciopus, we'd probably have to literally evolve it, i.e. in a rich simulate environment starting with simple artificial "organisms" that "feel" this environment and then getting progressively more complex.

But I do believe in Wolfram's "principle of computational equivalence", and thus that anything that can implement a turing machine can also implement any other complex system, including consciousness.

[whaaswijk]

I guess this is where we differ. I don't see a sufficient reason to believe that increasing computational capacity/complexity alone gives rise to consciousness. Moreover, I think there are common sense reasons to believe that consciousness is not substrate independent. Therefore, I don't see it as obvious that Turing completeness is sufficient for consciousness. For example, as someone else on this post has pointed out, a sufficiently complex water pipeline can implement a Turing machine. However, I doubt it would ever be conscious, no matter how large we make it. I think representing and processing information is orthogonal to experiencing.

[jbotz]

I think we do agree... "complexity alone" certainly will certainly not give rise to consciousness. Consciousness begins with feeling and separating the "I" from the "other"; I feel hunger, that there is food. That's why I said we'd have to evolve it in a simulated environment, one in which there are things for a nascent consciousness to feel. So in that sense yes, it depends on the substrate, but the substrate could be virtual, simulated on a powerful enough Turing machine.