Quote from: blablahblah on November 08, 2012, 04:28:26 PM

Quote from: ElectricMucus on November 08, 2012, 03:53:58 PM

Artificial Intelligence is a cybernetically impossible transformation. It's just not possible to create it, by definition.

Machines can be arbitrarily complex but they are defined in such a way that they depend on Man to control them. In computer science AI is used as a weasel word to describe mechanisms which attempt to solve problems using mathematical concepts which should, in theory enable the machine to compute solutions for problems it wouldn't have sufficient computational strength using other methods.
In transhumanism it refers to self-improving machines which again can not be constructed by definition. Every machine will still have a constraint defined by the parameters it is programmed even if it is able to construct copies of itself and use stochastic processes to fine-tune the parameters.

+1 I couldn't have put it better myself.

Quote from: grondilu on November 08, 2012, 03:30:53 PM

Quote from: blablahblah on November 08, 2012, 03:22:07 PM

As for 'the singularity', I call bullshit on that one too. It can't be done. Someone show me a compelling argument that it's theoretically possible for machines to have consciousness, and I will eat my words.

Why wouldn't it be possible? Why your brain would be so different from a machine?

As far as I can tell, I'm the only consciousness in existence, and "everything else" is just a product of my imagination in my little universe. At least with people, there is empirical evidence suggesting that they are capable of mirroring my feelings, sense of mercy or justice and many other human concepts.

Quote from: mobodick on November 08, 2012, 03:53:48 PM

Your definition of intelligence is too specific.
What you propably talk about is human intelligence.
And sure enough, human intelligence is so specific that we would need to recreate most structures of the brain to create such an intelligence.
But intelligence is a much broader concept.
Intelligence is best classified as an information system for dealing with the environment.
In that view even DNA molecules contain intelligence because they lead to specific manipulations of the environment.
Everything that manipulates the environment in a deliberate manner (acting on information) can be said to possess intelligence.
Human intelligence is just a very very specific case of intelligence.
In the case of an AI controling society, there is nothing that requires that AI to be concious or something like that.

Well, for the purposes of a central authority to run 'our' lives, why toy around with machines that are far simpler than humans? Why not use the best there is, i.e.: actual humans? Some might argue that it's a complex, rewarding job. Cheesy

Here's a hint.
The original problem was that we humans do a bad job.
Then why on earth would you want the replacement to act like a human?
It's a stupid projection of an ideal human onto a machine.
It's a crucial mistake to think that a better intelligence lies along the same lines as human intelligence.
But it is an easy mistake to make because human intelligence seems like the best kind of intelligence we know of.