On 9/15/06, Ben Goertzel <[EMAIL PROTECTED]> wrote:
However, I think that the introduction of provability into the discussion is largely a red herring. Current mathematics and science do not suffice to prove rigorous, nontrivial theorems about the behavior of complex systems in complex environments -- which is what theorems about action-based, let alone outcome-based, FAI would be. And it is quite likely that if we did have a revolution in complex computational systems theory, what it would tell us is why Friendliness canNOT be provably guaranteed -- rather than providing us with provable guarantees....
Right, it could be largely a red herring. In addition to reflecting upon the good points you made, to me it seems worth tinkering with the idea for a while that it's nontrivially absurd that Singularity.1 would be the superior optimization pressure. Judea Pearl thought and combinatorial optimization, among other entities, including their possible outgrowths, nonetheless seem to be a compelling pointer toward inductions of Friendliness feasibility being nonobviously unreasonable. Although, perhaps, metaactions won't come anywhere near being as rigorously guaranteed, thus reverting the theory back to a red herring, unfortunately. ----- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/[EMAIL PROTECTED]
