Silva, Guilherme Amorim Campos da2026-11-192026-11-192026https://repositorio.unialfa.com.br/handle/123456789/923This work consists of a systematic study on the algorithmic governance guidelines applied by the Court of Justice of the State of Goiás (TJGO). The incorporation of artificial intelligence (AI) by the Brazilian Judiciary boosts managerial efficiency but imposes complex regulatory and hermeneutic challenges. To mitigate the risks inherent to automation, algorithmic governance emerges, understood as the set of norms, processes, and institutional mechanisms aimed at submitting the design, implementation, and results of these technologies to democratic and constitutional control. Currently, there is a tension in the courts between the pressure for productivity and the material requirement for explainability and human-in-the-loop, vectors in which the social function of technology resides. In light of this, the central objective of this research is to evaluate how the mechanisms for the adoption, use, and oversight of AI seek a balance between operational efficiency and the safeguarding of fundamental procedural rights, notably due process of law, the motivation of decisions, the protection of personal data, and material equality, within the scope of the TJGO. The hypothesis guiding the investigation maintains that the legitimacy and effectiveness of this governance depend on the structuring of explainability and human supervision mechanisms (human-in-the-loop) that transcend mere formal validation or procedural triage and ensure the integrity of the decision-making foundation. Methodologically, the research has an applied nature and a qualitative approach, developed through bibliographic and documentary research, combined with an institutional case study, adopting the hypothetical-deductive method. The theoretical-normative analysis supports the formulated hypothesis, as it demonstrates that the sustainability of automation in the court requires the transition from a purely quantitative efficiency to a substantive governance. It is concluded, based on the proposition of ten institutional guidelines, that technological sovereignty and algorithmic governance do not constitute obstacles to innovation, but rather the indispensable condition for AI to remain subordinate to human jurisdiction and to realize the social function of Law in the digital age.ptinteligência artificial; governança algorítmica; direitos fundamentais; explicabilidade; supervisão humana.Diretrizes de governança algorítmica aplicadas pelo Tribunal de Justiça de Estado de Goiás (TJGO) na busca pelo equilíbrio entre a eficiência operacional e a salvaguarda dos direitos fundamentaisinfo:eu-repo/semantics/masterThesis