Legal implications of the use of artificial intelligence in election campaigns in Brazil

Abstract

With the concrete advent of generative artificial intelligence, societies have adopted new political communication strategies in national and local electoral processes, reshaping campaign dynamics and enhancing micro-targeting, automated content production, and the mass dissemination of misinformation. This study discusses the emergence of new vulnerabilities to the integrity of the electoral process, particularly in light of the level playing field among candidates and voters’ informational self-determination. Given this scenario, this study critically examines the legal implications of using artificial intelligence in electoral campaigns in Brazil, focusing on the civil, criminal, and electoral liability of candidates, parties, and digital platforms, as well as identifying existing regulatory gaps. In terms of methodology, this study adopts a qualitative, legal-dogmatic approach based on bibliographic and documentary research on the Brazilian legal system. Through an analysis of recent decisions by the Federal Supreme Court (STF), particularly in Cases 987 and 533 of general repercussion, as well as regulations issued by the Superior Electoral Court (TSE) regarding election advertising, misinformation, and the use of artificial intelligence, this study seeks to establish a critical dialogue with foreign regulatory initiatives, with a focus on the European Artificial Intelligence Regulation (AI Act) and its requirements for transparency, traceability, and the mitigation of systemic risks. Finally, the paper argues for the need for a regulatory framework that recognizes algorithmic informational power as a risk to democracy and establishes strengthened obligations regarding transparency, governance, and accountability, upholding the republican principles of the Democratic Rule of Law.