Desafios éticos e estruturais para usos e aplicações da inteligência artificial no poder judiciário brasileiro
Resumo
Resumo: A utilização da Inteligência Artificial (IA) vem crescendo exponencialmente e continua a evoluir de forma rápida, fazendo com que a revolução digital mude drasticamente. Grande parte dessa mudança se deve à explosão da IA generativa, que decorreu com o ChatGPT. As inovações tecnológicas estão remodelando as práticas jurídicas e legislativas na sociedade. Nesse contexto, o Poder Judiciário brasileiro se apresenta como um ambiente propício para a adoção da IA, considerando a alta carga de processos e a escassez de servidores judiciais. Contudo, tendo em vista os potenciais impactos e consequências da IA, existem desafios éticos e regulatórios. Dessa forma, o objetivo desta dissertação é verificar quais os parâmetros éticos essenciais presentes na literatura e nas diretrizes regulatórias para o controle ético da IA aplicada ao Poder Judiciário Brasileiro. Com base em análise bibliográfica, regulatória e legislativa existentes, a hipótese é que existem parâmetros éticos e regulatórios mínimos que podem balizar a aplicação e o uso da IA aplicada ao Poder Judiciário. Para isso, apresentam-se as principais definições de IA e a sua arquitetura. Em seguida, realiza-se uma taxonomia dos riscos envolvidos no uso da IA, como vieses, discriminação, desinformação, danos ambientais e socioeconômicos etc. Também investigamos as diretrizes éticas e regulatórias para aplicação da IA no Poder Judiciário, elencando os princípios éticos que mais se destacaram e analisando o atual panorama global da regulamentação da IA. Do mesmo modo, apresentam-se as principais aplicações de IA em sistemas de justiça do mundo, em especial, as utilizadas no Poder Judiciário Brasileiro. Por fim, analisamos os principais desafios que dificultam o uso e a implementação da IA no ambiente judiciário. As respostas fornecidas indicam: a falta de profissionais qualificados para trabalhar com IA; o excesso de trabalho enfrentado por esses servidores; a infraestrutura tecnológica insuficiente; a dificuldade na obtenção, armazenamento e uso responsável dos dados e; a resistência de servidores e magistrados na adoção de IA. Assim, verificou-se que são fundamentais investimentos na educação dos servidores e magistrados e na mãode- obra qualificada nos órgãos públicos. Também se identificou a necessidade de regulamentação da IA e a criação de padrões claros para sua implementação, a fim de tornar a definição e aplicação dos princípios éticos acessíveis e transparentes, visando promover um desenvolvimento ético da IA, centrado no ser humano, criando um ambiente seguro para investimentos na área e atendendo às necessidades dos cidadãos e jurisdicionados brasileiros. Abstract: The use of Artificial Intelligence (AI) has been growing exponentially and continues to evolve rapidly, causing the digital revolution to change drastically. Much of this change is due to a boom in generative AI, which emerged with ChatGPT. Technological innovations are reshaping legal and legislative practices in society. In this context, the Brazilian Judiciary presents itself as a suitable environment for the adoption of AI, considering the high workload of cases and the shortage of judicial staff. However, considering the potential impacts and consequences of AI, there are ethical and regulatory challenges. Thus, the objective of this dissertation is to verify which essential ethical parameters are present in the literature and regulatory guidelines for the ethical control of AI applied to the Brazilian Judiciary. Based on existing bibliographic, regulatory, and legislative analyses, the hypothesis is that there are minimum ethical and regulatory parameters that can guide the application and use of AI in the Brazilian Judiciary. To this end, the main definitions of AI and its architecture are presented. Next, a taxonomy of the risks involved in the use of AI is carried out, such as biases, discrimination, misinformation, environmental and socioeconomic damages, etc. We also investigate the ethical and regulatory guidelines for the application of AI in the Judiciary, listing the most prominent ethical principles and analyzing the current global panorama of AI regulation. Similarly, the main applications of AI in justice systems worldwide are presented, especially those used in the Judiciary. Finally, we analyze the main difficulties that delay the adoption and implementation of AI in the judicial environment. The responses provided by the courts indicate: the lack of qualified professionals to work with AI; the excessive workload faced by these staff; insufficient technological infrastructure; difficulty in obtaining, storing, and responsibly using data; and the resistance of staff and judges to the adoption of AI. Thus, it was found that investments in the education of staff and judges and in qualified labor in public bodies are essential. It was also identified the need for AI regulation and the creation of clear standards for its implementation, in order to make the definition and application of ethical principles accessible and transparent, aiming to promote the ethical development of AI, centered on the human being, creating a safe environment for investments in this field and meeting the needs of Brazilian citizens and litigants.
Collections
- Dissertações [683]