"A 'well-aligned' superintelligence will find this argument persuasive, they'll make sure"
This is the best argument for kicking EAs out of AI safety. Their moral compass is out of whack. We would be much better off with an AI trained by people who subscribed to Imago Dei