The Artificial Intelligence in the Sphere of the Administrative Law

Introduction

The extraordinary development of Artificial Intelligence (from now simply AI) in the context of administrative law, both in policy and in the main sectors of its organisational, administrative and jurisdictional activity, begins a new era. The term AI contains a series of notions: machine-learning, robotics, game theory, the development of complex algorithms, artificial neutral networks, etc. Administrative sector organizations has addressed the influence of AI on decision-making process, looking mainly at efficiency and rationalization. However, recent adoptions of AI have been challenged because of their discriminatory nature. As a result, questions emerged on the accountability of AI supported decision-making processes in the public sector. This process generates a lot of issues, with particular regard to the intermediation of the administrative provision. The so-called administrative relationship that is established between citizen and administration, rendered functional by the issue of a legislative measure, is affected by this process even before a provision is issued, not to mention the effects on the interaction between an administrative decision and its consequent effects on judicial review.

The automated administrative measure generates a series of advantages but also hides pitfalls not only for the public administration but also for the citizen himself. A computerised decision must not damage citizens but must be open to control. We are, therefore, faced with a double challenge, both technological and legal. A wise use of AI systems and programs should increase efficiency and performance in public service, while at the same time complying with constitutional and legal requirements and those norms in general that characterise administrative law. A legal challenge must restrict the use of new forms of technology and determine its application without placing unjustified and unnecessary obstacles in the way of technological development. AI will not only help the administrative assistant in the preliminary phases of the administrative process, concerning the admissibility, eligibility, validity and compatibility of applications, but also play a much more important role in the decision-making phases. However, this highly idealized and simplified vision risks undermining the fundamentals of administrative action inherent in the identification of the public interest in practice and the weighting in decision-making as well as the possible distortions of the administrative function.

The aim of this research is, therefore, to investigate, with a critical approach, the impact and the resulting effect that artificial intelligence has had in the sphere of Italian administrative law, also in relation to specific national circumstances within the EU area. On the 21 April 2021 European Commision proposed a new Union legislation to regulate AI. The Commission’s proposal marks a defining moment in the history of AI, since it will ultimately lead to the first comprehensive legislative measure globally containing binding rules on AI.