A Conceptual Framework for AI-based Decision Systems in Critical Infrastructures

Abstract-- The interaction between humans and AI in safety-critical systems presents a unique set of challenges that re main partially addressed by existing frameworks. These challen ges stem from the complex interplay of requirements for transparency, trust, and explainability, coupled with the neces sity for robust and safe decision-making. A framework that holistic ally integrates human and AI capabilities while addressing thes e concerns is notably required, bridging the critical gaps in designing, deploying, and maintaining safe and effective sys tems. This paper proposes a holistic conceptual framework for cri tical infrastructures by adopting an interdisciplinary approac h. It integrates traditionally distinct fields such as mathemati cs, decision theory, computer science, philosophy, psycholog y, and cognitive engineering and draws on specialized engineerin g domains, particularly energy, mobility, and aeronautics. Its flexibility is further demonstrated through a case study on power grid management. Artificial Intelligence (AI) is showing high potential to transform the management of critical infrastructures [1], tackling pressing challenges like climate change and the rising demand for energy and mobility systems while advancing strategic objectives such as energy transition and digi tal transformation. On the other hand, integrating AI in critic al sectors introduces significant challenges, many of which ar e already being addressed by emerging regulatory frameworks, such as the European Union AI Act. These frameworks emphasize the importance of safety, transparency, and adhe r-ence to ethical standards and principles to mitigate a wide range of risks, including technical, social, and environme ntal hazards associated with deploying AI in high-risk domains. Another key challenge lies in fostering effective human-AI collaboration.