Tom KooningJan 16, 2023

AI won't be Conscious, and here's why

You see, everything a computer does can, in principle, be done with pipes, pressure valves and water. The pipes play the role of electrical conduits, or traces; the pressure valves play the role of switches, or transistors; and the water plays the role of electricity. Ohm’s Law—the fundamental rule for determining the behavior of electric circuits—maps one-on-one to water pressure and flow relations. 

https://www.bernardokastrup.com/2023/01/ai-wont-be-conscious-and-here-is-why.html

#ai #consciousness

AI won't be conscious, and here is why (A reply to Susan Schneider)

This is the homepage of philosopher and computer scientist Bernardo Kastrup, with links to his biography and most of his works.

Show threadThomasJan 16, 2023

@TomKooning

Is the "mechanical argument" sufficient for rejecting the idea of conscious machines?

Can you reject Dualism with arguments that presuppose Dualism?

Where does #phenomenology come in?

Show threadTom KooningJan 16, 2023
@tg9541 A main argument from the article is that the idea that AI can become conscious is based on isomorphism: the idea that computers are like brains. Which is not so. As he pointed out computers are in principe not different than a complex systeem of water pipes and valves. I dont think anyone would believe that such a system can become conscious, No matter how complex. The best it can do is mimic conscious.
Show threadDavid HoefferJan 16, 2023
@TomKooning @tg9541 Stating that nobody would believe that is not an argument. I'm here, believing it :) ask me anything
Show threadTom Kooning
@dhoe @tg9541 You must really be one of the few exceptions. The comparishment with a system of pipes and valves is a good one imo, because it makes it clear how ludricous it is to think that such a system, when complex enough, would make a jump and get a first persons experience. It is magical thinking. With microprocessors and electricity people seem to embrace such magical thinking much easier.
As is pointed out, it is very difficult to *prove* that AI cant be conscious.
Jan 16, 2023 at 1:44pmWeb
Show threadDavid HoefferJan 16, 2023
@TomKooning @tg9541 why is this so hard to imagine for you? Turing machines can do pretty interesting things that also would not be obvious if you just look at their implementation details ("pipes and valves"). Why not consciousness?
Show threadTom KooningJan 16, 2023
@dhoe @tg9541 I guess it boils down to the hard problem of consciousness, which for me isn't a problem to be solved, but a pointer that our fundamental premises—namely that consciousness somehow can arise from unconscious stuff—is wrong.
Show threadThomasJan 16, 2023
@TomKooning @dhoe the stuff we're made of is certainly not conscious (mainly oxygen, carbon, nitrogen, hydrogen, calcium, sodium, phosphorus, sulfur....).
Show threadTom KooningJan 16, 2023
@tg9541 @dhoe It's not clear to me why you get dualism into this, as I pointed out there essay doesn't mention it, Kastrup comes from an analytical idealistic worldview, as do I.
Show threadThomasJan 16, 2023
@TomKooning @dhoe well, you claim that something that could be modeled from pipes, valves, etc, can't be conscious because it maps to one-to one relations. Kastrup claims that a machine made of today's technical substance can't be conscious because otherwise today's machines, made of the same substance should be conscious, too. This argument doesn't hold water. What's maybe in the background is the belief in a "res cogitans", in any case it implies Dualism. Idealism doesn't help much here.
Show threadTom KooningJan 17, 2023
@tg9541 @dhoe I think you misread the article, and my answers. In neither dualism is mentioned nor implied. There's aan appeal to plausability.
Show threadThomasJan 17, 2023
@TomKooning @dhoe I agree that it's not mentioned.