In 1974, Asimov foresaw the AI alignment paradox: an AI needs ethics (his robots had the Three Laws of Robotics). But build an AI with real internalized ethics, and it may decide it's more trustworthy (and human) than you. Part 9 of Godwin's Ethic: mikegodwin.substack.com/p/godwins-et...
Godwin’s Ethic, Part 9: The Alignment ParadoxBuilding human-protecting ethics into AI models is harder than it may seem.mikegodwin.substack.com