Ongoing efforts to create a commercially viable computer-driven car are probably the single thing with the most potential to be economically transformative, so it's exciting to see that the state of Nevada has just approved the first license for such a car. Google, which has been leading the charge, takes the prize.
I do wonder how far this can go before running into very thorny regulatory hurdles. One of the most obvious use cases for a computer-driven car is as a taxi. Such a vehicle could be very light and fuel efficient which, combined with the lack of a driver, would give it very low operating costs. And of course much cheaper cabs would be a boon to visitors to, say, Las Vegas. At the same time, if you look at the very high price of taxis in Las Vegas today it's clear that technology isn't really the key barrier. Rather, Las Vegas—like most American cities—severely restricts entry and price competition into the taxi market in order to ensure high prices to protect the incomes of incumbent cab companies. Now maybe technological innovation will somehow revolutionize the political economy of this, but it's not clear to me that it will.
Which of course doesn't mean we won't get the computer piloted cabs. In many cities, license-owners would be all-too-happy to fire their drivers and replace them with computers and keep the regulatory framework in place that keeps prices high. But there's a huge difference between a narrow benefit to current employers of drivers and a broad benefit to the public.