Algorithmic decision-making systems and legal responsibility: Reconstructing fault, agency, and accountability in the digital age

Abstract In the digital age, algorithmic decision-making systems have become central actors in shaping legal, economic, and social outcomes. From credit scoring and insurance risk assessments to welfare allocation, content moderation, and predictive policing, algorithmic systems increasingly determine individuals' access to rights, resources, and visibility. This transformation poses a fundamental challenge to classical legal concepts such as agency, fault, responsibility, and accountability, which are traditionally grounded in human intention and consciousness. While legal doctrine presumes a human decision-maker capable of intent, negligence, and moral judgment, algorithmic systems operate through opaque, data-driven processes that defy these assumptions. This article examines the legal nature of algorithmic decision-making systems from a doctrinal and legal-philosophical perspective. It argues that existing fault-based liability regimes are structurally inadequate to address algorithmic harms and proposes a reconceptualization of legal responsibility grounded in human dignity, transparency, and systemic accountability. The study ultimately advances the concept of "algorithmic accountability" as a normative framework capable of preserving the protective function of law in the digital era. Keywords: Algorithmic decision-making, legal responsibility, fault, human dignity, digital law, artificial intelligence