ARTIFICIAL INTELLIGENCE, EDUCATION AND SOCIAL INEQUALITIES: RISKS OF WIDENING EDUCATIONAL ASYMMETRIES AND REGULATORY PERSPECTIVES

Abstract

This investigation analyzes the socio-educational and legal impacts of artificial intelligence (AI) on the Brazilian public school ecosystem, aiming to diagnose the risks of worsening educational asymmetries and to propose guidelines for an Ethical Governance Framework aligned with the qualitative universalization of education. Adopting a qualitative, bibliographical, and documental approach, the study articulates the constitutional right to education, the theory of social inequalities, and the data protection framework to critically examine the digital incorporation of schools. The results reveal that algorithmic opacity, the biases of predictive systems, and the precariousness of infrastructure in the peripheries convert the promise of teaching personalization into a vector of social segmentation, reproducing structural exclusions of race, gender, and class. As a propositional contribution, five operational axes are systematized - federative governance, algorithmic impact assessment, digital inclusion, teacher training, and social control - unfolded into seven normative frameworks and two schematic representations. It is concluded that the subordination of automated tools to the democratic values of human dignity, material equality, and the supremacy of teacher agency is an indispensable precondition for the technological modernization of public schools to serve the emancipation of students and social justice.