The integrity of democratic elections worldwide is facing an unprecedented threat from sophisticated artificial intelligence tools capable of generating convincing fake multimedia content. Since 2018, electoral processes have been systematically undermined by disinformation campaigns, but the emergence of accessible deepfake technology in 2025 has dramatically accelerated this concerning trend.
Cybersecurity authority Rodolfo Avelino, speaking with Brasil de Fato’s radio station, characterized 2025 as a period of ‘impressive growth’ for AI manipulation tools. These advanced systems can now seamlessly replicate voices and visual appearances, creating fabricated content that challenges human discernment capabilities. The technological evolution has effectively blurred the line between authentic media and synthetic manipulations, creating unprecedented challenges for voters, journalists, and election authorities.
The Brazilian legislative response, known as the AI Regulation Bill (PL 2338, 2023), remains stalled in the Chamber of Deputies despite the escalating threat. Professor Avelino attributes this legislative paralysis to multiple factors, including significant corporate lobbying efforts regarding technology governance. The expert expressed minimal optimism for substantial regulatory progress in 2026, anticipating that more comprehensive policy measures are unlikely to reach voting stages.
The proposed legislation, under discussion for approximately five years, seeks to establish critical accountability frameworks including control mechanisms, impact responsibility protocols, and economic benefit distribution structures for AI technologies. Its continued stagnation leaves a dangerous regulatory vacuum during a period of rapidly advancing synthetic media capabilities.