Principles and Reasons Behind Automated Vehicle Decisions in Ethically Ambiguous Everyday Scenarios

2025-07-18

AI Summary

Autonomous driving systems frequently encounter ethically ambiguous everyday scenarios—such as lane-changing or social interactions—yet existing research disproportionately focuses on rare, high-stakes dilemmas, neglecting how human values are embedded in routine decision-making. Method: Through qualitative interviews and case studies, we systematically identify 13 distinct human-reason categories guiding autonomous behavior and propose a dynamic, four-layered decision framework spanning normative, strategic, tactical, and operational levels. Grounded in the “meaningful human control” theory—specifically its tracking condition—we derive actionable design principles. Contribution/Results: This empirically grounded, context-sensitive framework supports regulatory adaptability and multi-objective trade-offs while ensuring safety, thereby significantly enhancing ethical alignment and public acceptance of autonomous driving systems.

Technology Category

• Humans and AI: Learning Human Values and Preferences
• Philosophy and Ethics of AI: Safety, Robustness & Trustworthiness
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems

Abstract

Automated vehicles (AVs) increasingly encounter ethically ambiguous situations in everyday driving--scenarios involving conflicting human interests and lacking clearly optimal courses of action. While existing ethical models often focus on rare, high-stakes dilemmas (e.g., crash avoidance or trolley problems), routine decisions such as overtaking cyclists or navigating social interactions remain underexplored. This study addresses that gap by applying the tracking condition of Meaningful Human Control (MHC), which holds that AV behaviour should align with human reasons--defined as the values, intentions, and expectations that justify actions. We conducted qualitative interviews with 18 AV experts to identify the types of reasons that should inform AV manoeuvre planning. Thirteen categories of reasons emerged, organised across normative, strategic, tactical, and operational levels, and linked to the roles of relevant human agents. A case study on cyclist overtaking illustrates how these reasons interact in context, revealing a consistent prioritisation of safety, contextual flexibility regarding regulatory compliance, and nuanced trade-offs involving efficiency, comfort, and public acceptance. Based on these insights, we propose a principled conceptual framework for AV decision-making in routine, ethically ambiguous scenarios. The framework supports dynamic, human-aligned behaviour by prioritising safety, allowing pragmatic actions when strict legal adherence would undermine key values, and enabling constrained deviations when appropriately justified. This empirically grounded approach advances current guidance by offering actionable, context-sensitive design principles for ethically aligned AV systems.

Problem

• Addresses ethically ambiguous scenarios in AV everyday driving
• Explores routine decisions beyond rare high-stakes dilemmas
• Proposes framework for human-aligned AV decision-making

Innovation

• Applies Meaningful Human Control (MHC) framework
• Identifies 13 human reason categories for AV decisions
• Proposes safety-prioritized, context-flexible AV framework