In recent months, the advancement of artificial intelligence regulation in Brazil has ceased to be an abstract topic and has begun to have a direct impact on companies that already use (or intend to use) these tools in their operations.
Bill No. 2,338/2023, which establishes the regulatory framework for artificial intelligence, has already been approved by the Brazilian Senate and is currently under review by the Chamber of Deputies, pending subsequent presidential sanction. The bill proposes rules for the development, implementation, and responsible use of AI in Brazil.[1]
Among the central aspects of the proposal is the classification of AI systems according to their level of risk, with stricter requirements for applications considered “high-risk,” especially when they may affect fundamental rights.[2] The bill also provides for obligations related to governance, transparency, and bias mitigation, as well as ensuring user rights, such as explanations of automated decisions and the possibility of human review.[3]
In practice, this means that companies using AI in decision-making processes—such as credit analysis, pricing, or automated hiring—may be held liable for damages resulting from failures, biases, or lack of adequate oversight. The proposal itself adopts a liability model based on the level of risk of the system and the conduct of the agents involved.[4]
Moreover, even before the law is definitively enacted, the use of AI may already give rise to liability under existing legal frameworks, such as the Brazilian General Data Protection Law (LGPD)[5], particularly when it involves violations of rights or harm to third parties—an issue that has been widely discussed in legal scholarship and regulatory initiatives.[6]
From a contractual standpoint, relevant risks are already emerging:
- absence of specific clauses addressing AI use
- unclear allocation of liability between technology providers and users
- lack of auditability and traceability criteria for systems
The issue has moved beyond the technological sphere into the legal and strategic domain, and the decisions made now will directly impact companies’ future exposure.
Companies that act in advance tend to reduce litigation exposure and gain greater predictability in their operations. Recommended measures include:
- reviewing contracts with technology providers
- defining internal policies for AI use
- mapping risks in automated processes
- implementing human oversight mechanisms and corporate governance structures for AI
Once enacted, regulation will not create these risks—it will merely make more visible a responsibility that, in many cases, already exists.
This is therefore a moment for structuring—and companies that take action before final approval will be ahead of the curve.
[1] BRASIL. Ministério da Cultura. Senado Federal aprova marco regulatório da inteligência artificial. Disponível em: https://www.gov.br/cultura/pt-br/assuntos/noticias/senado-federal-aprova-marco-regulatorio-da-inteligencia-artificial. Acesso em: 16 abr. 2026.
[2] BRASIL. Câmara dos Deputados. Projeto regulamenta uso da inteligência artificial no Brasil. Disponível em: https://www.camara.leg.br/noticias/1159193-projeto-regulamenta-uso-da-inteligencia-artificial-no-brasil/. Acesso em: 16 abr. 2026.
[3] PL 2338/2023: regulamentação da inteligência artificial no Brasil. Jurídico AI. Disponível em: https://juridico.ai/direito-digital/pl-2338-2023-regulamentacao-ia-brasil/. Acesso em: 16 abr. 2026.
[4] RESUMO da PL 2.338/23, que dispõe sobre o uso da inteligência artificial no Brasil. Migalhas. Disponível em: https://www.migalhas.com.br/depeso/395117/pl-2-338-23-sobre-o-uso-da-inteligencia-artificial-no-brasil. Acesso em: 16 abr. 2026.
[5] BRASIL. Lei nº 13.709, de 14 de agosto de 2018. Lei Geral de Proteção de Dados Pessoais (LGPD). Disponível em: https://www.planalto.gov.br/ccivil_03/_ato2015-2018/2018/lei/l13709.htm. Acesso em: 16 abr. 2026.[6] INTELIGÊNCIA artificial no Brasil: PL 2338/2023, CNJ e ANPD. CriaAI. Disponível em: https://criaai.app.br/blog/regulamentacao-inteligencia-artificial-no-brasil/. Acesso em: 16 abr. 2026.