Legal Responsibility and Governance of AI in Judicial Systems: A Conceptual and Regulatory Framework for Algorithmic Decision-Making

This research develops a comprehensive legal and governance framework for artificial intelligence applications within judicial systems, addressing the growing reliance on algorithmic decision-making and its implications for legal responsibility. It situates AI within the transformation toward smart judiciary and examines the regulatory challenges arising from automation, opacity, and systemic complexity. The study argues that existing legal frameworks remain fragmented and insufficient to govern AI-driven judicial processes, particularly in relation to accountability, due process, and judicial independence. It identifies a critical gap between technological capabilities and normative legal structures, necessitating a reconstructed governance model. Adopting a doctrinal and comparative methodology, the research proposes a multi-layered governance framework based on risk classification, transparency, explainability, and human oversight. It integrates principles derived from emerging international regulatory models, including the EU AI Act, while adapting them to judicial contexts. The paper further explores the concept of algorithmic decision-making as a legally significant act, requiring formal recognition, procedural safeguards, and accountability mechanisms. It concludes by advocating for a hybrid regulatory approach that aligns innovation with the preservation of fundamental legal guarantees, contributing to the evolution of Legal Tech and smart justice systems.