4 ms·
This might be a naive question, but regarding “Let's imagine that we simulate the brain in all biological details on a supercomputer. Will that supercomputer b
by abotsis 6y ago
This might be a naive question, but regarding
“Let's imagine that we simulate the brain in all biological details on a supercomputer. Will that supercomputer be conscious?“
...wouldn’t the only reason for this be one of efficiency? That is, akin to how we can simulate one processor architecture on another.
Maybe the point is a simulation can never be efficient enough to become conscious (at least on a timescale we can grok)...?
- Koshkin 6y agoWell, consciousness is to a large degree a product of social interaction - in a way similar to a reflection requiring both a mirror and a light source; so, building a mirror alone is not enough. Therefore, if you "replicate" all neurons, synapses, etc. and emulate sensory inputs then you'll get a "literal copy" of a functioning brain of an actual person and then it will indeed be fully conscious.
- vladTheInhaler 6y agoScott Aaronson has an argument along roughly the same line regarding the Chinese room thought experiment. He points out that although it is usually framed as a "room" or a "box", the amount of space required to simulate a conversation of length n with just a lookup table is exponential. So in fact, the "room" would have to be colossally large for any reasonable n. There is a much more involved discussion of it, but the upshot is that the separation between polynomial and exponential time or space is a plausible place to draw the line between "really actually thinking and understanding" vs. "just a computation".