Artificial intelligence and speedy trial in the judiciary: Myth, reality or need? A case study in the Brazilian Supreme Court (STF)

Highlights

• AI is needed to deal with procedural delays.

• AI can impact three segments of speedy trial: support, pre-judgment, and judging.

• AI can reduce the cost and the time of judgment.

• AI can help legal professionals focus on the most valuable activities.

• Replacing legal professionals with AI is a myth.

Abstract

Justice has dealt with procedural delays for decades, contributing to the rise of Artificial Intelligence (AI) to face the problem. Therefore, it is essential to understand and analyze the implications of AI on speedy trial and how AI can accelerate judicial processes. To this end, we conducted a case study at the Supreme Court of Brazil (STF), collecting documents and conducting interviews for content analysis. The results consolidated a framework model that combines resources and impacts of AI on the velocity of legal decisions demonstrating how and what solutions contribute for judgment, pre-trial, and legal support. Therefore, it is believed that the gain in procedural speed with AI is not just a myth. It can become a reality, but with limitations, because there is a long way between the use of this technology in merely operational tasks and its use in complex activities such as evaluating processes in their entirety. For these future challenges, we highlight research proposals for more effective advances.

Introduction

For decades, the role of the Judiciary in several countries has been considered precarious (Bielen, Peeters, Marneffe, & Vereeck, 2017; French, 1933; Friesen & Information, 1984). In this context, one of the most debated issues is the speedy trial, which is related to the system's ability to respond to the demand by the Judiciary at an appropriate time (Ippoliti, Melcarne, & Ramello, 2014). Nevertheless, it has been observed high costs, long delays for the completion of proceedings and, consequently, procedural congestion (Gomes, Alves, & Silva, 2018).

Notwithstanding historical progress, delays and other problems in public services persist and are addressed continuously with Information and Communication Technology (ICT) (Dutta et al., 2019; Kum, Duncan, & Stewart, 2009). In this sense, Artificial Intelligence (AI) has been drawing attention (Chen, Guo, Gao, & Liang, 2019; de Sousa, De Melo, De Souza, Farias, & Gomes, 2019; Kankanhalli, Charalabidis, & Mellouli, 2019; Levmore & Fagan, 2019). There are courts in several countries that adhere to this technology (Alshahrani, Dennehy, & Mäntymäki, 2021; Julius, 2018).

Zeleznikow (2017), for example, presents the online dispute resolution systems (Online Dispute Resolution - ODR), which have proved to be a good option for improving the speed. On the other hand, Langbroek (2019) points out that AI does not appear to be the solution for reducing procedural stock or for increasing speed. It only reports that this technology can provide other types of benefits, such as improving the assertiveness of the evidence.

Thus, there are divergent hypotheses, but without checking the implications of AI on speedy trials and the factors associated with this phenomenon. Therefore, it is not clear whether the expected evolution occurs, which is corroborated by recent studies, which warn about the importance of understanding what the implications of new technologies for legal institutions are - with emphasis on AI (Andrade & Joia, 2012; Gomes et al., 2018; Iriberri, 2015; Omoteso, 2012; Wallace, 2017).

In addition to the whole context involving AI and speedy trials, de Sousa et al. (2019) draw attention to the need to investigate the use of AI in specific areas of the public sector, which constitutes a promising research opportunity. This study seeks to understand and analyze the use of AI in the Judiciary, more specifically, the contributions and limitations of this technology in the speedy trial.

To attend this objective, the Brazilian Superior Federal Court (STF) was selected for a case study, considering the analysis perspectives present in the literature. This study analyzes the STF's experience from the point of view of the implications of AI in the speedy trial. It also presents how AI has contributed to accelerating legal proceedings and obtaining a path for quantitative research that allows generalization.

The remainder of this paper is structured as follows. Section 2 presents the main concepts related to AI and speedy trial. Section 3 describes the research methods and techniques applied, detailing the case study and the literature review. Next, section 4 presents the results and discussions. In this section, it discusses the framework of the impact of AI on speedy trial. Finally, section 5 presents the conclusion, demonstrating the limitations of this study, and potential future researches.