Responsible AI System Development in Automotive Applications: A Framework

With the surge in adoption of artificial intelligence (AI) in automotive systems, especially Advanced Driver Assistance Systems (ADAS) and autonomous vehicles (AV), comes a proportional surge of AI-related incidents–several of which have ended in injuries and fatalities. These incidents all share a common deficiency: insufficient coverage towards safety, ethical, and/or legal requirements. Responsible AI (RAI) is an approach to developing AI-enabled systems that systematically takes such requirements into account. Existing international standards like ISO 21448:2022 (Safety of the Intended Functionality) do offer guidance in this regard but are not sufficient. Therefore, several industry-agnostic and domain-specific technical standards are emerging concurrently to address various RAI-related challenges, including but not limited to ISO/CD PAS 8800 for the integration of AI in automotive systems, ISO/IEC 5469:2024 for the integration of AI in functional safety, ISO/IEC 42001:2024 for implementation of an AI management system. This poses a significant risk of overwhelming organizations by forcing them to parse through overlaps and conflicts across these standards with their limited resources. In this paper, we explore a straightforward, concise, and high-level framework to harmonize guidance provided by existing AI standards with a focus on automotive applications. We show how this can streamline and standardize the integration of Responsible AI systems within the automotive industry and allow for the proactive development of safe, responsible, and ethical AI-enabled systems.