Uso da inteligência artificial para melhoria da produtividade e eficiência do judiciário

Abstract

Objective: The increasing adoption of Artificial Intelligence (AI) in the judiciary aims to address productivity challenges and resource constraints in the face of rising case volumes. However, this technological integration raises concerns about algorithmic bias, system opacity, privacy, and data protection. Metodology: This study presents a combined analysis consisting of a systematic literature review (2017–2024) and an exploratory investigation within the Court of Justice of the State of Sergipe (TJSE), in Brazil. Main Results: The systematic mapping followed a rigorous protocol, analyzing 33 articles selected from an initial pool of 752 studies found in scientific databases, focusing on AI approaches, challenges, productivity/efficiency benefits, and promising services. The exploratory study at TJSE involved an online survey with 56 participants (judges and staff from various departments), using a Likert scale to capture local perceptions regarding the impact of AI, expected benefits and challenges, and the prioritization of services to be enhanced through technology. Theoretical, methodological and management contributions: The review highlighted the predominant use of machine learning particularly deep learning and natural language processing with emphasis on the potential of models such as transformers and large language models. The most cited benefits include increased efficiency, productivity, and workload reduction, while key challenges relate to trust, human discretion, and the complexity of legal language. Societal contribution: Local perceptions confirmed global trends, emphasizing the value of information management automation and virtual legal assistance. It is concluded that, although AI offers significant advantages for the judiciary, its implementation requires ethical caution and transparency to ensure a fair, trustworthy, and human-centered judicial system.