Kant is my co-driver: philosophy and driverless cars

Above and beyond the pure engineering, efficiency and sustainability goals of autonomous vehicles lies the safety goals. Autonomous vehicles have been shown to be safer than human-driven vehicles. The Google car has been on the streets for years now, and has seen only a few accidents, and almost all of these have been the fault of the human-driven other vehicle. However, as we transition to a driverless fleet, questions of fault and ethics move to the forefront. An autonomous vehicle cannot bear fault as it cannot claim agency, however, in the case of an accident is the fault that of the programmer? The manufacturer? The owner? Further (and more controversially) how should an autonomous vehicle react in a no-win situation, where all outcomes lead to injury for at least one human, be they passenger or pedestrian? To read Alan Ryan's discussion of this dilemma, see: https://www.timeshighereducation.co.uk/opinion/philosophy-and-driverless-cars-kant-is-my-my-co-driver