THE DOUBLE HELIX OF GENERATIVE ARTIFICIAL INTELLIGENCE: VECTOR AND ANTIDOTE TO DISINFORMATION IN BRAZIL

Abstract

The rise of Generative Artificial Intelligence (GAI) has dramatically reshaped the information and communication ecosystem, establishing a central paradox: the same technology that enables the creation and dissemination of disinformation at unprecedented scale and sophistication also offers the most promising tools for its detection and mitigation. This review article critically analyzes this duality, which resembles a double helix, focusing on the Brazilian context. The methodology adopted was a systematic literature review, encompassing academic articles, technical reports, and legislative documents, to map current trends and challenges. The results demonstrate, on the one hand, how technologies such as Generative Adversarial Networks (GANs) and Large Language Models (LLMs) are instrumentalized to produce deepfakes and false narratives, impacting democratic processes and public trust. On the other hand, the potential of AI as a tool for automated fact-checking and analysis of disinformation patterns is highlighted. The analysis delves into the Brazilian regulatory debate through a comparative study between Bill No. 2338/2023 and the European Union's AI Act, revealing similarities and implementation challenges. The conclusion is that safe navigation in the new hybrid infosphere demands a multifaceted approach that integrates robust technological regulation, ethical adaptation of communication practices, and a massive investment in media and algorithmic literacy for society.

DOI: https://doi.org/10.56238/sevened2025.029-091