From: Kate Riley
> ....My problem with this notion
> of AI is that it is inherently circular, in that ultimately, the only way
> could know that the AI is phenominally more intelligent than any of us is
> for a being of phenominally high intelligence to tell us so.
Kate, I know that there are many posters to this list that are phenomenally more intelligent than myself. And there is even a higher tier composed of phenomenally higher intelligences here that would be happy to tell me that this is the case. Testing, prediction and retrodiction are a few of the tools that could confirm this 'suspicion' I have about their superior intelligence. You view the argument as being circular. I see it as more 'spiral' -like in representation.
> ....if the AI agreed with everything the human populace agreed with,
> it would be pretty useless to us as a Power.
I do not see AI and Power as equivalent. As Eliezer pointed out, the Cro-Magnons were the last enormous evolutionary step in the hominid progression. Achievement of AI is one of the possible approaches that could culminate in the next step. And from there, exponential/geometric evolution might progress to this 'Power'....or it might not. Hominid evolution has been rather boringly flat, as of late, it's time to take that next leap, dontcha think?
> Therefore, if the AI decided that the human species should be obliterated,
> would be justified in calling it a bad judgment call and taking arms
> against it.
"True warfare in which large rival armies fight to the death is known only
in man and in social insects." Dawkins: The Selfish Gene. One can only hope
that the AI has not 'inherited' this aspect of socialization that we seem