I don’t either but I find “the best of all possible worlds” concept very interesting along with the related notion God could not possibly create a world that was anything other than “the most awesome of all possible worlds”—given the predicates traditionally ascribed to God.
You can take this is a reductio ad absurdam of the notion of God as many do. But presumably the task of friendly AI (at its most benevolent) must be to perform (or figure out) what actions should be taken to promote awesomeness?
More extravagantly, the task of friendly AI is to ‘build God.’
I don’t agree with Leibniz, but I do find his “best of all possible worlds” concept really useful for talking about what utilitarians try to do.
I don’t either but I find “the best of all possible worlds” concept very interesting along with the related notion God could not possibly create a world that was anything other than “the most awesome of all possible worlds”—given the predicates traditionally ascribed to God.
You can take this is a reductio ad absurdam of the notion of God as many do. But presumably the task of friendly AI (at its most benevolent) must be to perform (or figure out) what actions should be taken to promote awesomeness?
More extravagantly, the task of friendly AI is to ‘build God.’
(And get Her right this time :-)