logoalt Hacker News

c1ccccc1today at 7:24 PM0 repliesview on HN

Just a terminology note: Alignment does not mean the AI will help its owner kill people. (Indeed, an AI aligned to value human life would generally try to prevent murders.) The word for an AI that follows all instructions of its owner, as that owner intended them to be understood, is "corrigible" or "controllable".