Personal Data Protection and Algorithmic Auditing: Inter-American Standards on the Use of AI by Tax Authorities

Abstract

This article critically examines the use of artificial intelligence (AI) systems in tax enforcement processes and their impact on the fundamental rights to personal data protection, privacy, and due process. Likewise, it analyzes mass data processing, the construction of tax risk profiles, and decision-making based on algorithmic models as contemporary expressions of state surveillance and punitive power. To this end, a legal-dogmatic analysis of a constitutional and conventional nature is conducted on the Mexican regulatory system that governs personal data protection, public administration actions, and procedural guarantees. Based on the constitutional block, risks of decision-making opacity, algorithmic biases, undue expansion of administrative powers, and the relativization of procedural guarantees are identified. Finally, the paper proposes legal criteria to guarantee technological enforcement that respects human rights and their exercise in the Mexican context, grounded in the pro persona principle, the Political Constitution of the United Mexican States, the American Convention on Human Rights, and the jurisprudence of the Inter- American Court of Human Rights.