Artificial intelligence and judicial systems : opportunities and challenges at the intersection of AI and courts of justice

Abstract

The adoption of Artificial Intelligence (AI) in the judicial systems is a phenomenon that has been gaining strength in recent years and presents opportunities and challenges for the achievement of fundamental rights. In order to offer an up to date material on this regard for Human Rights defenders, this work provides an overview of the leading technologies involved in AI, Machine Learning, and Big Data, starting with their applications in two case studies: China e Estonia. While the former already experiences the presence of AI-judges in its cyber courts focused on internet-related cases, the latter is developing an ongoing project that foresees AI's implementation as a decision-maker in small cases in the first instance of its judicial system. These cases are followed by a comparative analysis of recent researches that applied NLP and Machine Learning to carry out predictive experiments on judicial decisions, with an accuracy rate of up to 90%. This panorama serves as a support to understand the main opportunities at the intersection of AI and justice — such as the speedup of processes, bringing more efficiency, and reducing the backlog of cases, which hypothetically guarantees for individuals a more accessible and of a better quality judicial procedures. Still, some risks arise at the same proportion — such as the threat of a repetition of discrimination patterns, the perpetuation of inequalities, and loss of human job positions. To address these challenges, legal instruments and other mechanisms to regulate the use of AI emerge worldwide, some of them targeting judicial systems in particular.