Racismo algorítmico em iinteligências artificiais: o enviesamento tecnológico a serviço da discriminação

Abstract

This study investigates the presence of algorithmic racism in artificial intelligence and its role in perpetuating discrimination. The general objective is to examine how algorithmic racism manifests in artificial intelligence, contributing to the perpetuation of racial discrimination. The specific objectives are to identify how algorithmic racism is presented in artificial intelligence; to verify the ways in which technologies can reproduce racism; and to investigate how ethical failures in artificial intelligence algorithms, especially those related to algorithmic racism, are being addressed by national legislation. The chosen methodology is bibliographic research with a qualitative approach, using databases such as the Digital Library of Theses and Dissertations (BDTD); Google Scholar; the CAPES Journal Portal and Scielo Brazil to gather articles, dissertations and theses. The results show that algorithmic racism occurs systematically, driven by discriminatory biases present in the data. This negatively impacts areas such as facial recognition, where algorithms reinforce historical prejudices and marginalize minorities. The study highlights the importance of ethical principles in artificial intelligence and the need for robust legislation to combat these discriminatory practices, promoting equity and social justice.