would anyone like to team up to create Artificial Intelligence based on personal interest?

Question

The type Im referring to is a program that can fight off the side of humanity which is increasingly turning evil. Humans just cannot last long enough to stand against the changing mentality of this evolving world. Need to create something which does the right thing. (although I realize right and wrong are relative). Something which can distinguish and defend.

Orichalc, you advice of seeking something worthy doesnt make sense. anything which borders too close to real Intelligence and mirrors it is interchangable. cannot say which one is real and which one is AI. in that case, i do not see any purpose in trying to create any sort of AI at all since there is no goal or purpose.

to Dr.D : then i guess its an individual project after all.

*correction Dr.R

Answer 1

You're trying to find a high-tech replacement for religion. You admit that right and wrong are relative, which I assume to mean relative to the standards of the individual or society, yet you appear willing to leave such decisions to a machine. This sounds risky since people are decidedly inconsistent in their application of moral principles, yet a machine could be expected to be consistent. For example, "the greatest good for the greatest number" sounds like a good moral principle, but to a machine (for which time is an abstraction) programmed with such a principle, sacrificing billions now in a mass eugenics and euthanasia program to create a super-healthy race and environment some 100 years from now might seem like a good thing. However, if you leave the possibility for humans to veto such a plan, then you've done nothing to avoid man's vulnerability to the corruption of the vetoing powers. I'm afraid there is no quick fix.

Answer 2

I would team up for a project more worthy of attention. Seek for an article "Creating Friendly AI" or CFAI.html somewhere in Google Search, to learn something.