The rapid deployment of autonomous artificial intelligence (AI) agents on cloud infrastructure has enabled scalable and intelligent automation across a wide array of industries, from healthcare diagnostics to financial fraud detection. However, as these AI systems operate with increasing autonomy and interact dynamically with users and data in cloud environments, the traditional boundaries of responsibility become blurred. When such agents malfunction or cause unintended consequences, it is often unclear who is to be held accountable: the developers, the cloud service providers, the end-users, or the AI agents themselves. This study aims to explore and clarify the complex landscape of responsibility attribution in autonomous AI systems deployed on the cloud. By integrating ethical theory, legal principles, and system architecture perspectives, the paper investigates real-world case studies and develops a responsibility mapping framework. The results highlight critical gaps in existing legal and ethical structures and propose a multi-stakeholder attribution model to enhance transparency, trust, and accountability in cloud-AI ecosystems.
Emmanuel Idowu (Wed,) studied this question.