Artificial intelligence is not really human, so we should always avoid conditions like “reliable AI” that not merely humanize AI but also indicate a level of dependability that simply isn't going to exist. Ricardo and Cansu both desire the term “responsible AI” whi