The Principle of Transparency in Automated Decision-Making Systems – French Legal Framework

ABSTRACT This paper examines the emerging legal framework governing algorithmic transparency in French public Administration. It focuses on three key dimensions of transparency: the right of access to public algorithms, the right of individuals to be informed when algorithmic systems are involved in administrative decision-making, and the requirement of explainability. While these rights are increasingly recognized in both French and European law, their effective implementation remains uncertain. The paper highlights the legal, technical, and organizational challenges that limit their practical effectiveness and argues that the main difficulty lies not in the recognition of transparency obligations, but in their enforcement and day-to-day application by public authorities.

KEYWORDS Algorithmic transparency - Automated decision-making - Public algorithms - Right to explanation - French public administration

TABLE OF CONTENTS 1. Introduction. – 2. The Right of Access to Public Algorithms. – 2.1. Legal Recognition. – 2.2. Legal Limits. – 3. The Right of a Person to Be Informed That They Have Been Processed by an Algorithm. – 4. The Requirement of Explainability. – 4.1. The Regulatory and Judicial Framework. – 4.1.1. The Law First Requires AI Systems to Be Interpretable. – 4.1.2. Individuals Also Have the Right to Receive Explanations Regarding Algorithmic Decisions Made About Them. – 4.2. The Practical Challenges and Limits of Legal Implementation. – 4.3. Potential Avenues for Addressing These Challenges. – 5. Conclusion.