Algorithmic justice, understood as the application of algorithms and artificial intelligence within judicial systems, has already raised numerous concerns regarding constitutional safeguards. While in European legal systems, including the Italian context, it remains confined to experimental initiatives without yet achieving widespread structural implementation, certain foreign experiences, particularly that of the United States, have underscored the urgency of examining the phenomenon from a constitutional law perspective. Following an introductory chapter that delineates the scope of the inquiry and outlines the methodology employed, the first part of this work, utilizing an empirical and inductive approach, is devoted to an analysis of the current state of algorithmic justice, with particular emphasis on the U.S. experience concerning the use of risk assessment tools in criminal proceedings and the experiments with so-called predictive justice in European legal systems. The second part of the study focuses on the regulatory frameworks governing artificial intelligence and algorithmic justice in both the United States and Europe, with special attention to recent developments in Europe such as the Artificial Intelligence Act (European Union) and the Framework Convention on Artificial Intelligence (Council of Europe). Having outlined the current state of art and the applicable regulatory frameworks, the third part of the thesis reflects on the impact of algorithmic tools on judicial decision-making and their implications for constitutional guarantees, particularly those related to the right to a fair trial within the framework of European constitutional law. Drawing on the extensive U.S. jurisprudence on the compatibility of algorithms in judicial proceedings with judicial discretion and due process of law, the analysis centers on those constitutional guarantees that appear most challenged by the use of algorithms in judicial decision- making. These include judicial independence and impartiality, the right to defense, equality of arms, and the right to an effective remedy. Finally, acknowledging the issues at stake, the concluding part of the work, also in light of recent regulatory solutions adopted in Europe, seeks to reflect on the concept of algorithmic fair trial, considering both the limitations and opportunities of algorithmic justice and the guarantees necessary to accompany such a profound transformation.
La giustizia algoritmica, intesa come l’impiego di algoritmi e dell’intelligenza artificiale all’interno dei sistemi giudiziari, ha già sollevato numerosi interrogativi sul piano delle garanzie costituzionali. Ancorché negli ordinamenti europei e in quello italiano sia ad oggi confinata a iniziative sperimentali e non abbia ancora raggiunto un impiego strutturale, alcune esperienze straniere, soprattutto quella statunitense, hanno già mostrato l’urgenza di analizzare il fenomeno da una prospettiva costituzionalistica. Dopo un primo capitolo introduttivo in cui viene delimitato il perimetro dell’indagine e indicata la metodologia utilizzata nella ricerca, una prima parte del lavoro, secondo un metodo empirico e induttivo, è dedicata all’analisi dello stato dell’arte della giustizia algoritmica, con un approfondimento particolare sull’esperienza statunitense di impiego di strumenti di previsione del rischio di recidiva nel procedimento penale e sugli esperimenti della cd. giustizia predittiva negli ordinamenti europei. La seconda parte mira invece alla ricognizione degli approcci regolatori dell’intelligenza artificiale e della giustizia algoritmica negli Stati Uniti e in Europa, nel cui ambito sono di recente stati approvati il Regolamento sull’intelligenza artificiale (Unione Europea) e la Convenzione quadro sull’intelligenza artificiale (Consiglio d’Europa). Chiariti stato dell’arte e regolazione, la terza parte della tesi riflette sull’impatto dell’impiego degli algoritmi sulla decisione del giudice e sulle ricadute che ciò comporta sulle garanzie costituzionali, in particolar modo su quelle legate al giusto processo nella prospettiva costituzionale europea. Muovendo dagli insegnamenti della vasta giurisprudenza americana sulla compatibilità degli algoritmi impiegati nel processo con la discrezionalità del giudice e con il due process of law, l’indagine si focalizza su quelle garanzie costituzionali che, ad oggi, appaiono più “sotto pressione” in conseguenza dell’utilizzo di algoritmi da parte del giudice, tra cui l’indipendenza e l’imparzialità dello stesso, il diritto di difesa, la parità delle armi e il diritto a un ricorso effettivo. Infine, presa coscienza delle problematiche in gioco, l’ultima parte del lavoro tenta, anche alla luce delle recenti soluzioni normative adottate in Europa, di ragionare sul giusto processo algoritmico, guardando i limiti e le opportunità della giustizia algoritmica e le guarentigie necessarie per accompagnare tale profondo cambiamento.
LA GIUSTIZIA ALGORITMICA ALLA PROVA DELLE GARANZIE COSTITUZIONALI
VALSECCHI, ANDREA
2024
Abstract
Algorithmic justice, understood as the application of algorithms and artificial intelligence within judicial systems, has already raised numerous concerns regarding constitutional safeguards. While in European legal systems, including the Italian context, it remains confined to experimental initiatives without yet achieving widespread structural implementation, certain foreign experiences, particularly that of the United States, have underscored the urgency of examining the phenomenon from a constitutional law perspective. Following an introductory chapter that delineates the scope of the inquiry and outlines the methodology employed, the first part of this work, utilizing an empirical and inductive approach, is devoted to an analysis of the current state of algorithmic justice, with particular emphasis on the U.S. experience concerning the use of risk assessment tools in criminal proceedings and the experiments with so-called predictive justice in European legal systems. The second part of the study focuses on the regulatory frameworks governing artificial intelligence and algorithmic justice in both the United States and Europe, with special attention to recent developments in Europe such as the Artificial Intelligence Act (European Union) and the Framework Convention on Artificial Intelligence (Council of Europe). Having outlined the current state of art and the applicable regulatory frameworks, the third part of the thesis reflects on the impact of algorithmic tools on judicial decision-making and their implications for constitutional guarantees, particularly those related to the right to a fair trial within the framework of European constitutional law. Drawing on the extensive U.S. jurisprudence on the compatibility of algorithms in judicial proceedings with judicial discretion and due process of law, the analysis centers on those constitutional guarantees that appear most challenged by the use of algorithms in judicial decision- making. These include judicial independence and impartiality, the right to defense, equality of arms, and the right to an effective remedy. Finally, acknowledging the issues at stake, the concluding part of the work, also in light of recent regulatory solutions adopted in Europe, seeks to reflect on the concept of algorithmic fair trial, considering both the limitations and opportunities of algorithmic justice and the guarantees necessary to accompany such a profound transformation.File | Dimensione | Formato | |
---|---|---|---|
phd_unimi_R13277.pdf
accesso aperto
Dimensione
3.67 MB
Formato
Adobe PDF
|
3.67 MB | Adobe PDF | Visualizza/Apri |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/184562
URN:NBN:IT:UNIMI-184562