PROCEDURAL JUSTICE IN THE ERA OF ALGORITHMS: A CRITICAL EVALUATION OF AI GOVERNANCE AND DIGITAL RIGHTS

Abstract

Today, AI is being added to government and business so quickly that international law cannot keep up, creating a risky situation for basic rights and freedoms. We argue that while AI offers unmatched efficiency, it simultaneously acts as a conduit for systemic risks, primarily through the mechanisms of algorithmic bias, opacity, and unauthorized data exploitation. This paper highlights critical disparities, such as commercial facial recognition systems exhibiting error rates of up to 35% for darker-skinned women compared to less than 1% for lighter-skinned men. Furthermore, we examine how algorithms in U.S. hospitals were found to be less likely to refer Black patients to healthcare programs than equally ill white counterparts. Our core thesis suggests that existing legal instruments—conceived for a pre-digital, human-centric era—lack the technical specificity and agility required to regulate autonomous black box systems. We adopt a multidisciplinary methodology, synthesizing doctrinal legal analysis with empirical evaluations of AI's impact on vulnerable populations. By critically examining the limitations of current regimes like the GDPR, we highlight the widening gap between technological capability and legal accountability. We conclude that without mandatory human-rights impact assessments and strong liability standards for developers, AI risks entrench historical inequalities. Lastly, we propose a shift toward adaptive regulation that prioritizes the right to an explanation to preserve human dignity in the digital age.