Ética na inteligência artificial no serviço público brasileiro - uma revisão sistemática da literatura sobre princípios, aplicações e desafios

Resumo

Objective – This study conducts a systematic literature review on the application of artificial intelligence (AI) in the public sector, focusing on the ethical principles of transparency, fairness, accountability, privacy, and non-maleficence. It aims to understand how these principles are addressed globally and to identify the main challenges to their implementation in the Brazilian context. Methodology – The research followed the PRISMA protocol, analyzing 1,354 studies retrieved from academic databases such as CrossRef, CORE, and PubMed. Text mining and extraction algorithms were used to classify articles based on keywords related to ethical principles. All stages of the process included manual validation. Results – The literature highlights recurring topics such as algorithmic bias and fairness but reveals gaps in the practical application of ethical principles. In the Brazilian scenario, the findings indicate a lack of specific regulation, low transparency in algorithmic systems, and difficulties in ethical oversight. Regarding Bill No. 2,338/2023, which proposes Brazil’s AI legal framework, the study concludes that, despite being a normative advancement, it still lacks clear mechanisms for implementation. Limitations – The main limitations include the restriction to publications in English and Portuguese, the inability to fully read all analyzed texts, and limited access to certain scientific repositories. Practical implications – The findings can support policymakers, researchers, and institutions in developing more robust guidelines for the ethical use of AI in Brazil’s public sector. Social contributions – By identifying concrete challenges in the ethical adoption of AI, this study contributes to the debate on algorithmic governance and public accountability, fostering greater transparency and fairness in the use of such technologies. Originality – This is one of the first studies to combine automated systematic review methods with an analysis of the ethical challenges of AI specifically applied to the Brazilian public sector.

Ficha do documento

Tipo

Dissertação

Ano

2025

Instituição

Fundação Getulio Vargas

Fonte

Repositório da FGV

Idioma

Português

Acesso

Acesso aberto

Identificador

oai:repositorio.fgv.br:10438/37183

Temas

Gestão PúblicaAdministração PúblicaTecnologiaGovernança

Palavras-chave

Inteligência artificialÉticaSetor públicoGovernançaRegulamentaçãoRevisão sistemática de literaturaBrasilBrasileiroArtificial intelligenceEthics, public sectorGovernanceRegulationSystematic literature reviewBrazilianAdministração públicaServiço públicoGovernança pública