Between hallucinations and innovations

the challenge of responsible implementation of Generative AI in the Justice System

Authors

DOI:

https://doi.org/10.12662/2447-6641oj.v23i44.p1-26.2025

Keywords:

Generative Artificial Intelligence, risks and potential, Judiciary, human protagonism, governance

Abstract

Context: The potential of Generative Artificial Intelligence (GAI) in the Judiciary, while generating great expectations regarding increased efficiency, has motivated significant concerns, reproducing the perplexities that have always permeated humanity's relationship with innovation, notably technological innovation.

Goal: The investigation that gave rise to this article had two objectives: to gather and systematize some concepts and knowledge about GAI, its characteristics, evolution, and language models; and to assess how GAI can be employed in the context of the Judiciary, with proper risk management and harnessing of its potential. The hypothesis to be explored is that the risks of Generative Artificial Intelligence (GAI) in the justice system lie not only in its misuse but also in its non-use.

Method: The research is bibliographic and documentary. It begins with a historical overview of computing and artificial intelligence – AI, presents general notions about AI, Machine Learning, and Generative AI, then explores the risks of ignoring the characteristics and limitations of GAI. It then examines how this innovation is being absorbed by the Judiciary, identifying its potential benefits.

Results: The results reveal that there are risks of hallucinations in responses, reproduction of biases, and inadequate handling of sensitive data, but also that these risks can be managed through proper knowledge of GAI’s functioning and limitations, governance guided by ethics, and human protagonism. They also show that, given the volume of data and information that circulate and are generated by the Judiciary, the use of GAI is becoming a condition of possibility in knowledge management for an adequate and coherent judicial performance.

Conclusions: The hypothesis was confirmed, identifying that the risks of GAI in the Judiciary lie both in not knowing how to use it and in not using it at al.

Author Biographies

Cíntia Menezes Brunetta, FADISP/UNIALFA

Doutora em Direito pela Faculdade Autônoma de Direito (Fadisp). Conselheira do Conselho Nacional do Ministério Público (CNMP). Juíza federal do Tribunal Regional Federal da 5 Região. Professora da Graduação e dos Programas de Mestrado e Doutorado da Fadisp. Pesquisadora da Fundação Nacional de Desenvolvimento do Ensino Superior Particular (Funadesp). Coordenadora Adjunta do Mestrado Interinstitucional Fadisp/Ejug. Formadora de magistrados e formadora de formadores da Enfam e de outras escolas de magistratura e do Ministério Público. Fortaleza - CE - BR.

Taís Schilling Ferraz, Enfam - Escola Nacional de Formação e Aperfeiçoamento de Magistrados

Doutora em Ciências Criminais e Mestre em Direito pela PUCRS. Desembargadora do TRF4. Professora do quadro permanente e vice-coordenadora do Programa de amestrado da ENFAM e membro do Centro Nacional de Inteligência da JF. Brasília - DF - BR.

Alisson Carvalho de Alencar, FADISP/UNIALFA

Pós-doutorando pela Universidade de São Paulo (USP); Doutor em Direito pela Universidade de Salamanca, em convênio com a Faculdade Autônoma de Direito (2020); Mestre em Administração Pública pela Escola Brasileira de Administração Pública da Fundação Getúlio Vargas (2015). Procurador Geral de Contas do Ministério Público de Contas do Estado de Mato Grosso. Cuiabá - MT - BR. 

Downloads

Published

2025-10-09

How to Cite

BRUNETTA, Cíntia Menezes; FERRAZ, Taís Schilling; ALENCAR, Alisson Carvalho de. Between hallucinations and innovations: the challenge of responsible implementation of Generative AI in the Justice System . Revista Opinião Jurídica (Fortaleza), Fortaleza, v. 23, n. 44, p. 1–26, 2025. DOI: 10.12662/2447-6641oj.v23i44.p1-26.2025. Disponível em: https://periodicos.unichristus.edu.br/opiniaojuridica/article/view/5913. Acesso em: 17 sep. 2026.

Issue

Vol. 23 No. 44 (2025)

Section

Artigos Originais

License

Copyright (c) 2025 Cíntia Menezes Brunetta, Taís Schilling Ferraz, Alisson Carvalho de Alencar

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Copyright and Licensing

According to the current submission policy, authors retain the rights to their works and grant the Revista Opinião Jurídica the right of first publication, with commercial rights reserved by the publisher under the terms of the non-commercial license used. Revista Opinião Jurídica uses a Creative Commons license. The works published are under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC-BY-NC-SA).

This license enables reusers to distribute, remix, adapt, and build upon the material in any medium or format for noncommercial purposes only, and only so long as attribution is given to the creator. If you remix, adapt, or build upon the material, you must license the modified material under identical terms. CC BY-NC-SA includes the following elements:

 BY: credit must be given to the creator.
 NC: Only noncommercial uses of the work are permitted.
 SA: Adaptations must be shared under the same terms.

RESPONSIBILITY FOR THE CONTENT

By submitting an article, the author(s) declare to have sole responsibility for the content of the piece and is(are), therefore, responsible for any judicial or extrajudicial measures referring to it.

1. In case of joint authorship, all authors are considered collectively responsible, except when proved otherwise.