I consider myself a pretty reckless optimist when it comes to the general topic of artificial intelligence and the specific topic of driverless cars. In particular, I figure that true driverless cars will be here within a decade or so. But apparently I'm not being aggressive at all. In fact, I'm just an old fogey. Via Matt Yglesias, here's a projection from a recent Morgan Stanley research report:
Damn! "Limited" driver substitution by next year. And full-on driverless cars within four or five years. I sure hope the researcher who wrote this knows something I don't. If this is all really in the works, it's pretty awesome.