AI APPLICATIONS IN FACIAL RECOGNITION AND PORTRAIT OPTIMIZATION

Abstract

This study analyzes the applications of artificial intelligence in facial recognition and portrait optimization, highlighting technological advances, ethical, legal, and social implications, as well as interdisciplinary possibilities. The objective was to understand how deep learning algorithms, especially convolutional neural networks, have transformed biometrics and digital authentication systems while raising debates on privacy, algorithmic bias, and regulation. The justification lies in the need to critically examine a technology that, while offering greater efficiency and practicality, may also generate risks of excessive surveillance and violations of fundamental rights. The methodology followed a qualitative, exploratory, and bibliographic approach, grounded in Lakatos and Gil, using documentary analysis of Brazilian scientific articles in PDF. The results showed that although current models achieve high levels of accuracy in controlled environments, they still face limitations in real-world scenarios and generate social distrust. It was also observed that portrait optimization is relevant in several areas, such as medicine, psychology, advertising, and digital security, demonstrating the interdisciplinary nature of the subject. It is concluded that the evolution of facial recognition must be tied to regulatory mechanisms, transparency, and social responsibility, in order to align technological innovation with the preservation of rights and human values.