The convergence of Generative AI (GenAI) and Functional Safety (FuSa) under ISO 26262 is reshaping Advanced Driver Assistance Systems (ADAS) by merging adaptive intelligence with stringent safety assurance. GenAI significantly improves perception, object classification, pedestrian intent prediction, and risk assessment, empowering vehicles to navigate highly dynamic, real-world environments. Furthermore, GenAI-driven synthetic data generation accelerates ADAS validation by enabling exhaustive testing of rare and safety-critical scenarios not covered by conventional datasets.
However, the integration of GenAI into safety-critical automotive systems introduces major challenges. Its probabilistic nature, limited transparency, and vulnerability to adversarial inputs directly conflict with ISO 26262 principles that demand determinism, traceability, and fault isolation.
This paper proposes a co-engineering framework that combines GenAI’s learning-based adaptability with deterministic safety mechanisms to ensure real-time, fail-operational behavior. The use of Explainable AI (XAI) is explored as a means to enhance model interpretability and support functional safety argumentation. Additionally, architectural strategies such as functional redundancy, runtime monitoring, and formal verification techniques are presented to address the unpredictability of AI models.
By offering a development roadmap, this study enables safe and certifiable integration of GenAI into ADAS. It highlights the importance of cross-functional collaboration among AI developers, safety engineers, and system architects, along with the adoption of robust validation methodologies. This paper contributes to advancing the next generation of intelligent, trustworthy, and safety-compliant automotive systems that can leverage the full potential of GenAI without compromising reliability.