← Global AI Policy Radar · Methodology
Jurisdiction Baseline · AI × Policy

AI Regulation in Brazil.

At a glance
AI-Specific Law
No comprehensive federal AI law currently in force
Core Legal Framework
Privacy, consumer and internet laws govern AI
Platform Duties
2026 rules require risk controls and content safeguards
Child Protection
Child-accessible AI products face enhanced safety duties

As of August 6, 2026, Brazil has no comprehensive federal AI-specific law in force. AI is instead governed mainly by the LGPD, Brazil’s data protection law, consumer law, the Marco Civil da Internet, Brazil’s internet framework law, and 2026 platform rules covering systemic risks, illegal content, and AI-generated intimate deepfakes; the ECA Digital, Brazil’s child online safety law, also applies to products likely to be used by children or adolescents. Enforcement is active and increasing, but Brazil has no general AI licensing or conformity assessment system.

In force today6 instruments
Lei Geral de Proteção de Dados Pessoais (LGPD) in force
Applies whenever an AI system processes personal data, including for training, profiling, recommendation, fraud detection, biometrics or automated decisions. Companies must establish a lawful basis, observe purpose and necessity limits, secure the processing, respect data-subject rights and manage international transfers and incidents; the ANPD can investigate and impose sanctions. ([planalto.gov.br](https://www.planalto.gov.br/ccivil_03/_ato2015-2018/2018/lei/l13709.htm?utm_source=openai))
Read the full text ↗ · 2018
Data processingAI / automated decisionsSafety & security
Marco Civil da Internet in force
Sets baseline duties for internet connection and application providers, including privacy, security, logging, user rights and rules for intermediary liability. It is particularly important for AI products that host, rank, recommend or moderate user-generated content. ([planalto.gov.br](https://www.planalto.gov.br/ccivil_03/_ato2011-2014/2014/lei/_leis2014.htm?utm_source=openai))
Read the full text ↗ · 2014
Data processingContent moderationSafety & security
Unlock the full baseline

Get the remaining 4 instruments and the founder FAQ. Free: subscribe once and every briefing on this device unlocks.

Already subscribed? Unlock →

Decreto nº 12.975/2026 in force
Updates the Marco Civil regulation and imposes operational duties on internet application providers, especially platforms, concerning transparency, risk monitoring, prevention and mitigation of systemic failures, illegal content, fraud and paid promotion. It matters for AI platforms that distribute or amplify user content, although it is not a general AI law. ([planalto.gov.br](https://www.planalto.gov.br/ccivil_03/_ato2023-2026/2026/decreto/d12975.htm?utm_source=openai))
Read the full text ↗ · 2026
Content moderationDisclosureSafety & security
Decreto nº 12.976/2026 in forceAI-specific
Creates binding platform duties on digital violence against women, expressly covering AI-generated or AI-manipulated intimate content. Covered platforms must implement safeguards against generating or modifying intimate deepfakes, operate reporting channels and remove specified content within set time limits. ([planalto.gov.br](https://planalto.gov.br/ccivil_03/_ato2023-2026/2026/decreto/d12976.htm?utm_source=openai))
Read the full text ↗ · 2026
Content moderationSafety & securityDisclosure
Estatuto Digital da Criança e do Adolescente in force
Applies to technology products or services directed at, or likely to be accessed by, children and adolescents. AI services in scope must provide elevated privacy, safety and security, prevention, information and parental-control measures, with ANPD oversight and implementing rules under Decreto nº 12.880/2026. ([planalto.gov.br](https://www.planalto.gov.br/ccivil_03/_ato2023-2026/2025/lei/l15211.htm?utm_source=openai))
Read the full text ↗ · 2025
MinorsData processingSafety & security
Código de Defesa do Consumidor in force
Applies to AI products and services offered to consumers, including SaaS, chatbots, recommendation systems and automated customer service. It prohibits misleading or abusive practices, requires clear information and supports liability for defective or unsafe services, regardless of whether the harm was caused by an algorithm or a human. ([planalto.gov.br](https://www.planalto.gov.br/ccivil_03/leis/l8078compilado.htm?sf40763266=1&utm_source=openai))
Read the full text ↗ · 1990
Consumer protectionDisclosureSafety & security
Also worth noting
  • PL 6.237/2025, the government's proposed framework for a national AI governance system and allocation of regulatory powers, is not yet a law. ([planalto.gov.br](https://www.planalto.gov.br/ccivil_03/projetos/ato_2023_2026/2025/pl/pl-6237.htm?utm_source=openai))
  • PL 2.338/2023, the Senate's broader AI regulatory framework, remains pending and should not be treated as binding law. ([www25.senado.leg.br](https://www25.senado.leg.br/web/atividade/materias/-/materia/157233/pdf?_materia_WAR_atividadeportlet_conversion=isca-agentes-ia&utm_source=openai))
  • ANPD regulatory work on AI and emerging technologies is active, including planned monitoring, guidance and possible rules under the LGPD. ([gov.br](https://www.gov.br/anpd/pt-br/assuntos/noticias/anpd-publica-agenda-regulatoria-2025-2026?utm_source=openai))
Pending and recent legislation8 measures
BR PL 4881/2026 Algorithmic Integrity and Consumer Protection in Artificial Intelligence
Introduced · 2026-08-04
This bill establishes governance, security, transparency, responsibility, compliance, and consumer-protection duties for the use of artificial intelligence systems in consumer relationships. It creates an Algorithmic Integrity Legal Regime for Consumer Relations and amends Brazil
BR PL 2873/2026 PL 2873/2026, Establishes the National Policy for Digital Inclusion and Artificial Intelligence for Family Farming, AGROFAMÍLIA 5.0
Advancing · 2026-08-03
The bill creates a national policy to promote digital inclusion and the use of artificial intelligence in family farming. It likely addresses access to digital technologies and AI-enabled tools for family farmers.
BR PL 2850/2026 National Technological Career Transition Policy
Advancing · 2026-08-03
Establishes the National Technological Professional Transition Policy (PNTPT) to prepare, protect, and retrain workers in response to the effects of automation, artificial intelligence, and new technologies on the labor market.
BR PL 2019/2026 PL 2019/2026, National Policy for Labor Transition, Human Oversight, and Limits on Automated Replacement in the Labor Market
Advancing · 2026-08-03
Establishes a national policy for labor transition and minimum protections against automation-driven job displacement, including requirements for human oversight and limits on automated substitution in the labor market. It also sets related parameters for managing the effects of
BR PL 2885/2026 PL 2885/2026, Establishes the National Policy for Digital Inclusion, Technological Innovation, and Artificial Intelligence for Family Farmi
Introduced · 2026-07-29
The bill establishes a national policy to promote digital inclusion, technological innovation, and artificial intelligence in family farming. It appears focused on expanding access to digital tools and AI-enabled technologies in the family-farming sector.
BR PL 4701/2026 PL 4701/2026, Bill Ensuring Citizens’ Right to Human Review and a Minimum Explanation of Administrative Decisions Made by Automated Systems
Introduced · 2026-07-17
The bill would give citizens the right to human review and a minimum explanation for administrative decisions made by automated systems or standardized protocols. It directly addresses automated decision-making in public administration.
BR PL 4695/2026 Criminalizing Nonconsensual Synthetic Sexual Content
Introduced · 2026-07-17
Prohibits the creation, solicitation, dissemination, and facilitation of non-consensual synthetic sexual content depicting an identifiable person. It establishes criminal offenses and aggravating factors, and imposes prevention, removal, administrative-sanction, and emergency-res
BR PL 4692/2026 Citizen Rights in Automated Government Decisions
Introduced · 2026-07-17
The bill establishes citizens’ safeguards for the use of automated systems by Brazil’s public administration, including the right to challenge data used in automated decisions and a ban on permanent automatic administrative blocks without human review.

Live from the Global AI Policy Radar.

Founder FAQAI law, regulators, what applies, what is next, penalties
Is there an AI-specific law in force here?

No. As of August 6, 2026, Brazil has no comprehensive federal AI-specific law in force. There is also no general AI licensing or conformity-assessment regime.

Who are the key regulators for AI?

The key enforcement bodies are the ANPD, consumer authorities, courts and sector regulators. The ANPD oversees data protection and has specific oversight under the ECA Digital.

Which rules apply to AI systems today?

The LGPD applies when AI systems process personal data, including for training, profiling, recommendations, fraud detection, biometrics or automated decisions. The Marco Civil da Internet, consumer law, the ECA Digital and the 2026 platform rules on systemic risks, illegal content, fraud, paid promotion and AI-generated intimate deepfakes may also apply, depending on the product and use case.

What is coming next, and when?

The source does not identify a further comprehensive AI law or provide a future timetable. The ECA Digital is already in force, with implementing rules under Decreto nº 12.880/2026, and the platform rules in Decretos nº 12.975/2026 and nº 12.976/2026 are already in force.

What are the enforcement and penalty risks?

Enforcement is real and becoming more active, led by the ANPD, consumer authorities, courts and sector regulators. The ANPD can investigate and impose sanctions under the LGPD, while consumer law supports liability for misleading or abusive practices and defective or unsafe services. The source does not specify particular penalty amounts.

Not legal advice. For educational purposes only. AI-researched against official sources (2026-08-06), links verified.

Need a company-specific assessment? Explore Nomos →