As a conscious being I prove my existence by engaging with external stimuli…
…Bots can pass the Turing test, but passing the Turing test doesn’t necessarily guarantee consciousness.
This is part of the problem. We don’t have a consistent definition for consciousness anymore than we have a definition for AGI. (AGI can, by reading the instructions, build flat-packed furniture, or make coffee, but would a bot that could do these things be AGI?)
We assume the people we talk to are conscious, but then they could be Turing complete bots, or a Chinese room, or a p-zombie. You’ve essentially argued that you cannot demonstrate to us that you are actually conscious, only that you seem so convincingly.
Similarly, if I were to argue that I’m not conscious, but an advanced iteration of an AI program practicing speaking from a private lab in Sacramento California, and in fact, have no life beyond going online and pretending to be a person, you’d have no way of establishing this as true or false.
So appealing to consciousness is useless on account that we can’t actually say what it is. Again, we don’t have any edge cases of anything that is nearly conscious and appears to be, but isn’t, or something that is conscious but only barely. We assume that anything we can engage as human is, often leading to peculiar results like Sophia, the Robot-yet-Saudi-citizen that isn’t even convincingly sophisticated.
I’d argue that we want to be more than a material chain reaction, to the point that we’re frightened of considering the bare minimums that we would need to be to be convincingly ourselves.
As a conscious being I prove my existence by engaging with external stimuli…
…Bots can pass the Turing test, but passing the Turing test doesn’t necessarily guarantee consciousness.
This is part of the problem. We don’t have a consistent definition for consciousness anymore than we have a definition for AGI. (AGI can, by reading the instructions, build flat-packed furniture, or make coffee, but would a bot that could do these things be AGI?)
We assume the people we talk to are conscious, but then they could be Turing complete bots, or a Chinese room, or a p-zombie. You’ve essentially argued that you cannot demonstrate to us that you are actually conscious, only that you seem so convincingly.
Similarly, if I were to argue that I’m not conscious, but an advanced iteration of an AI program practicing speaking from a private lab in Sacramento California, and in fact, have no life beyond going online and pretending to be a person, you’d have no way of establishing this as true or false.
So appealing to consciousness is useless on account that we can’t actually say what it is. Again, we don’t have any edge cases of anything that is nearly conscious and appears to be, but isn’t, or something that is conscious but only barely. We assume that anything we can engage as human is, often leading to peculiar results like Sophia, the Robot-yet-Saudi-citizen that isn’t even convincingly sophisticated.
I’d argue that we want to be more than a material chain reaction, to the point that we’re frightened of considering the bare minimums that we would need to be to be convincingly ourselves.