Brazil's AI Bill (PL 2338) Awaiting Final Congressional Approval

The Current Status of Brazil's AI Bill: PL 2338

Brazil's Artificial Intelligence bill, known as PL 2338, is frequently discussed as if it were already enacted law. However, this is a critical distinction: the risk tiers, prohibitions, and proposed regulatory bodies that are often attributed to Brazilian law currently exist only within a text that has been approved by one chamber of Congress, not both.

PL 2338/2023 was first introduced in the Federal Senate in May 2023. Its foundation was built upon the report of a commission of jurists tasked with drafting a comprehensive AI regulatory framework to replace earlier, less detailed proposals. Following extensive committee deliberations throughout 2024, the Senate plenary gave its approval to the bill on December 10, 2024. Subsequently, it was sent to the Chamber of Deputies for further examination. As of the current date, the bill has not been enacted into law; it has successfully passed through one legislative chamber and awaits the decision of the other.

This legislative process is dynamic, and the bill's journey through Congress is ongoing. The distinction between a bill passed by one chamber and an enacted law is significant, particularly concerning the practical application of its provisions. Many summaries and analyses of PL 2338 describe its effects as if they are already in force, which can lead to misunderstandings about the current legal landscape for AI in Brazil.

Understanding the Legislative Process

The journey of PL 2338 through Brazil's Congress highlights the country's structured legislative process. Introduced in the Senate, it underwent a rigorous committee review phase. This phase is crucial for refining the bill, incorporating feedback from various stakeholders, and addressing potential concerns. The commission of jurists played a pivotal role in shaping the initial draft, drawing on extensive legal expertise to create a robust framework.

Upon approval by the Senate, the bill transitioned to the Chamber of Deputies. Here, it is being reviewed by a special committee. This committee's role is to conduct a thorough examination of the bill, potentially proposing amendments before it is brought to a floor vote. The fact that it is being handled by a special committee, rather than being fast-tracked for a plenary vote, suggests a deliberate approach to scrutinizing the legislation.

The current status means that the detailed provisions concerning AI, such as the classification of AI systems into different risk tiers, specific prohibitions on certain AI applications, and the establishment of a regulatory authority, are not yet legally binding. They represent the proposed framework that Congress is considering. Until the bill receives final approval from both the Chamber of Deputies and the President of Brazil, it remains a proposal, not an enacted law.

Key Provisions Under Consideration

While PL 2338 is not yet law, understanding its proposed provisions offers insight into Brazil's approach to AI governance. The bill aims to establish a comprehensive legal framework for the development, deployment, and use of AI systems within the country. Central to its structure is a risk-based approach, categorizing AI systems based on their potential to cause harm.

The proposed risk tiers are designed to impose varying levels of regulatory scrutiny. High-risk AI systems, those that could potentially infringe on fundamental rights or pose significant safety concerns, would be subject to the most stringent requirements. These might include mandatory impact assessments, continuous monitoring, and enhanced transparency obligations. Conversely, low-risk AI systems would face fewer regulatory burdens, allowing for innovation and adoption.

Furthermore, the bill outlines specific prohibitions for AI applications deemed unacceptable. These typically target systems that pose an undue risk to fundamental rights, such as certain forms of social scoring or AI used for manipulative purposes. The intent is to draw clear lines on what constitutes ethically and legally problematic AI deployment.

A key element of the proposed legislation is the establishment of a regulatory body. This entity would be responsible for overseeing AI development and deployment, enforcing the law, and issuing guidelines. The precise structure and powers of this regulator are subjects of ongoing debate and refinement within the legislative process.

Implications of the Bill's Status

The fact that PL 2338 is still a bill, not a law, has significant implications for businesses, developers, and researchers operating in or targeting the Brazilian market. Any discussions or implementations based on the bill's provisions should be framed with the understanding that these are not yet legal mandates. Companies are currently operating under existing general legal frameworks, such as data protection laws (like LGPD) and consumer protection statutes, rather than specific AI regulations.

For developers, this means that while it is prudent to anticipate future regulatory requirements and design AI systems with ethical considerations and potential compliance in mind, immediate adherence to the specific risk tiers or prohibitions outlined in PL 2338 is not legally required. The timeline for enactment remains uncertain, dependent on the legislative schedule and the outcome of deliberations in the Chamber of Deputies.

The distinction is not merely academic. It affects how companies plan their AI strategies, invest in compliance measures, and communicate their AI practices. Relying on the current draft of PL 2338 as a definitive legal guide would be premature. Instead, stakeholders should monitor the legislative process closely, paying attention to potential amendments and the final text as it progresses toward potential enactment.

Looking Ahead

The progress of PL 2338 through Brazil's Congress signifies a proactive step toward regulating artificial intelligence. The detailed nature of the bill, as it stands after Senate approval, suggests a comprehensive approach to AI governance that balances innovation with the protection of fundamental rights. However, its journey is not yet complete.

The focus now shifts to the Chamber of Deputies. Their review will determine the final form of the legislation. It is possible that amendments will be introduced, altering the scope or specific provisions of the bill. Stakeholders, including industry players, civil society organizations, and legal experts, will be closely watching these deliberations.

Until PL 2338 is officially enacted into law, its provisions serve as a strong indicator of the direction Brazil intends to take in governing AI. Businesses and developers should remain informed about its progress and prepare for eventual compliance, but understand that the current legal reality in Brazil is not yet shaped by this specific AI bill.