Brazil’s Lower House Advances EU-Inspired AI Bill, Imposing New Compliance Rules on Global Tech

Brazil’s evolving regulatory landscape for Artificial Intelligence took a decisive turn as the Communications Commission (CCom) of the Chamber of Deputies approved Bill PL 2.688/2025, which institutes a new risk-based regulatory framework for the development and use of AI in the country. The move signals a high-stakes compliance environment for US-based and global technology companies operating in Brazil, one of the world's largest consumer markets, by adopting a framework explicitly inspired by the European Union’s landmark AI Act.

The proposed legislation is a significant step in the country's multi-track approach to AI governance. While a broader framework (PL 2.338/2023) moves through the Chamber after approval in the Senate, PL 2.688/2025 focuses specifically on operational obligations and the definition of illicit conduct related to AI use, introducing civil and penal transparency norms. This approach places Brazil at the forefront of AI regulation in Latin America, focusing on a human-rights-centered model where systems are classified based on their potential impact on fundamental rights.

For platforms and developers, the new bill mandates a substantial increase in operational and algorithmic transparency. Core requirements include the identification and tracking of artificial content—such as deepfakes—the provision for human review in automated decisions that carry a potential impact on individual rights, and comprehensive transparency and auditability for AI systems. Furthermore, the framework alters the country’s foundational internet law, the Marco Civil da Internet (Civil Rights Framework for the Internet), to strengthen collaboration with authorities on cybercrimes. This change comes as Brazil's Supreme Court has already eroded the Marco Civil's protections for platforms against liability for third-party content, making the operational rules in the new AI bill a critical layer of compliance for major social media and platform companies.

The advancement of the bill through the Chamber of Deputies, the lower house of Brazil's National Congress in Brasília, means the text must still pass through other technical committees, including Science, Technology and Innovation; Finance and Taxation; and Constitution, Justice and Citizenship, before a vote in the Plenary. If it becomes law, it will establish a clear, enforceable standard that defines acceptable risk for AI deployment in key sectors, placing the onus on developers and operators to document and mitigate potential harm to Brazilian users.

What it touches The bill creates new and potentially costly compliance requirements for global technology companies, including those with significant operations in Brazil that rely on AI for content moderation, recommendation engines, and customer interaction. The increased demand for auditability and the potential for civil and criminal liability for misuse translate into a substantial operational risk and compliance expense for platforms, impacting the local subsidiaries and operations of companies that make up the global technology sector.