As the act of driving becomes increasingly automated, vehicles will encounter situations where different objectives of safety, mobility, and legality will come into conflict. These situations require a vehicle to compare relative values of different entities and objectives, where the action of the vehicle has a moral component. While discussion of these scenarios often focuses on the “trolley problem” thought experiment, these types of life-or-death moral dilemmas may be rare in practice. This article identifies four far more common examples of routine driving that require decisions with some level of ethical reasoning about how to distribute risk. These scenarios may be useful for automated vehicle developers in assessing vehicle safety and responding to potential future regulations, as well as for regulators in developing performance requirements.
If the inline PDF is not rendering correctly, you can download the PDF file here.