The Economist just published this (https://t.co/BfgD2pKt9W )- and the question is whether the main premise is a bad thing. We treat it as desirable for AI to help us become smarter at doing things, but why wouldn't we also use AI to help us be more considerate about our values?