BrazilAI regulation

Regulation Status

• Brazil has no comprehensive AI law yet in force, but a formal legislative framework is well advanced: the Senate approved Bill No. 2338/2023 in December 2024, which then moved to the Chamber of Deputies. The bill sets a risk-based classification, prohibits 'excessive risk' AI practices, mandates impact assessments for 'high-risk' systems, and grants creators a copyright opt-out from AI training datasets.
• Brazil therefore has no comprehensive, binding AI-specific law in force; it relies on sectoral rules, ANPD enforcement and guidance, and a national AI plan while the merged 2338/6237 framework advances through Congress.
• Pending the statute, regulators apply existing law to AI: the ANPD has run AI public consultation and a regulatory sandbox and supports the government's AI bill, the competition authority CADE and the ANPD/MPF/Senacon have pursued Meta and X's Grok over AI data use and non-consensual deepfakes, and sectoral bodies (the electoral court on election deepfakes, the medical council CFM) have issued AI rules.
• In December 2025 a complementary governance bill, PL 6237/2025, was introduced, creating a National System for the Development, Regulation and Governance of AI (SIA) with a new Brazilian Council for AI (CBIA) and the data protection authority ANPD as central coordinator; the government has moved to merge PL 2338/2023 and PL 6237/2025 into a unified framework, with Brazil still ending 2025 without a definitive AI law.