> Humans often give coherent _rationalizations_ of their actions, which is slightly different than an explanation. At least those rationalizations are logicall

> Humans often give coherent _rationalizations_ of their actions, which is slightly different than an explanation. At least those rationalizations are logically plausible as explanations; they very well could be (usually) the actual explanation. What's now called AI cannot, AIUI, convincingly come up with any "explanation" of its reasoning, simply because it hasn't done any actual reasoning. I don't know if most fans or current "AI" really can't see the difference, or are just pretending. (If they really can't, one can't help wonder if they are themselves current "AIs"...)