Oscar Capdeferro

Automated controls against fraud and corruption in Spain

Introduction

Automated controls against fraud and corruption in spain. Study automated controls against fraud and corruption in Spain, analyzing AI challenges (transparency, errors, unreliability). Proposes national legal safeguards beyond the EU AI Act.

Abstract

This paper examines the current state of affairs concerning the regulation of automated administrative activities related to inspection, supervision, and control in Spain, drawing on existing legal frameworks and selected case studies. It critically reflects on three key challenges associated with the use of AI and algorithmic systems in oversight functions: the lack of transparency, the occurrence of errors, and the potential unreliability or insufficient robustness of system outputs. The study explores several possible solutions and concludes by emphasizing the need to incorporate specific legal safeguards into national legal systems, in light of the insufficient protections currently provided by the EU AI Act for most systems employed in supervisory functions.

Review

This paper tackles a highly relevant and critical subject: the integration of automated administrative activities, specifically AI and algorithmic systems, into inspection, supervision, and control functions aimed at combating fraud and corruption in Spain. Drawing on existing legal frameworks and selected case studies, the study provides a timely examination of the current regulatory landscape. Its primary contribution lies in offering a critical reflection on three fundamental challenges associated with these technologies in an oversight context: the persistent lack of transparency, the potential for errors, and the overarching concern of system unreliability or insufficient robustness, all of which are crucial considerations for public trust and administrative justice. A key strength of this work appears to be its analytical depth, moving beyond mere description to a critical evaluation of the practical and ethical implications of AI in public administration. The mentioned methodology, incorporating both legal frameworks and specific case studies, suggests a well-rounded approach that grounds theoretical concerns in real-world applications. By exploring "several possible solutions," the paper indicates a constructive outlook, aiming to offer practical recommendations rather than simply identifying problems, thereby enhancing its potential utility for policymakers and practitioners grappling with similar challenges globally. The paper concludes with a particularly significant call to action, emphasizing the urgent need for incorporating specific legal safeguards into national legal systems. This recommendation stems from the astute observation that the protections offered by the EU AI Act are currently insufficient for the majority of systems employed in supervisory functions, underscoring a critical regulatory gap. This finding has substantial implications not only for Spain but for other jurisdictions within the EU and beyond, positioning the work as an important voice in the ongoing discourse on responsible AI governance in the public sector. The study thus provides a strong foundation for future research and policy development aimed at ensuring the ethical and effective deployment of automated controls.

Full Text

Comments

You need to be logged in to post a comment.