Ethical frameworks for automated vehicles: a systematic analysis and design

Abstract As automated vehicles (AVs) transition from research environments to public roads, the absence of a robust pluralistic meta-ethical framework to guide their decision-making presents a significant challenge in ensuring their actions align with societal values. This paper proposes a comprehensive, multi-layered meta-ethical framework tailored to AVs, combining principles from Augmented Utilitarianism and biomedical ethics. We begin by examining various existing ethical frameworks, highlighting the relevance of biomedical ethics due to its focus on minimizing harm and decision-making under uncertainty. Augmented Utilitarianism serves as the foundational layer of our framework, defining the moral space through the perspectives of the agent, action, patient, and experiencer—representing society—while maintaining flexibility for real-world applications. Within this moral space, biomedical ethics provide the guiding attributes for ethical decision-making. Our proposed framework seeks to ensure comprehensive ethical coverage, addressing harm minimization, fairness, and societal welfare, while offering the adaptability needed for unforeseen situations. Given that such a framework does not yet exist, this work is essential for guiding the ethical operation of AVs. Furthermore, we propose the refinement of this framework through an experimental process involving key stakeholders to ensure it remains aligned with societal values and technological advancements. This research underscores the importance of an adaptable and iterative approach to AV ethics, ensuring the technology evolves responsibly in alignment with societal values.