The comparison with cyclists isn't completely apt, though. I can't really think of an empirical benefit for human-driven cars over computer-driven (assuming the

The comparison with cyclists isn't completely apt, though. I can't really think of an empirical benefit for human-driven cars over computer-driven (assuming there's a manual mode that can be activated outside the cities and off freeways), but governments have incentives to encourage cycling (smaller parking footprint (municipal government spends less on parking spaces), less wear on the road (lowers maintenance costs), health benefits for the riders (lower public health costs), less fuel usage (improves energy independence), etc.). I'll not make an argument about whether those factors outweigh the inconvenience of motorists dodging cyclists on the road, but I don't think there are any analogous reasons it might be a good idea to allow human-driven cars on the road with autonomous cars.