It seems unrealistic to think that any machine of sufficient intelligence will not have bias (relative to some perfectly imagined, imperfectly designed human sy

It seems unrealistic to think that any machine of sufficient intelligence will not have bias (relative to some perfectly imagined, imperfectly designed human system). Any goal assumes some a priori good or set of norms over which to optimize. Philosophers, politicians and priests are still working on that one.