Much is made these days about artificial intelligence or AI particularly as it may be the savior for what ails self-driving vehicle research.
But what AI really amounts to today is a bunch of basic rules enhanced by algorithms that can adjust their output based on some other Bayesian probabilities. I've glossed over some details, but you get the picture: it's not artificial intelligence in the sense of duplicating what we humans do, think.
So what? We may not need true AI after all. Here's a great presentation by John Searle given to a gaggle of Googlers that puts it in all in perspective.