Lessons from Brazil and Peru on how to build AI-ready institutions across Latin America

Lessons from Brazil and Peru on how to build AI-ready institutions across Latin America

November 19, 2026 • 5:17 pm ET Olivia Torres, Angel Ortega Latin America’s artificial intelligence (AI) ambitions are beginning to collide with the limits of its institutions. Countries in the region are developing AI regulatory frameworks, many drawing on European regulatory traditions, from data protection to emerging risk-based AI rules. But imported models must be translated into clear authority, operational capacity, and accountability at home. While still at an early stage of AI governance, Brazil and Peru have become regional leaders in regulation, pioneering different approaches. Brazil has begun strengthening its institutions while its general AI framework remains in development. Peru has become one of the first Latin American countries to pass a general AI law coupled with implementation mechanisms. Together, these cases offer two models for other countries in the region, organized around three questions: Is authority clear? Which body leads implementation, and how are responsibilities divided? Can institutions absorb new mandates? Do they have sufficient personnel, expertise, funding, and administrative capacity? Can institutions detect and correct failure? Are transparency, review, participation, and accountability mechanisms developing alongside deployment? Lessons from Brazil: build institutions, even before laws Brazil is building the institutional architecture for AI governance while its general framework remains before a special committee in the Chamber of Deputies. No vote date has been confirmed. The Senate approved a bill on December 10, 2024 regulating the use, development, and deployment of AI. An attached proposal would expand the role of Brazil’s National Data Protection Authority (ANPD), making it the default AI regulator alongside sectoral regulators. On February 25, 2026, another law turned ANPD into a full regulatory agency with greater technical, administrative, and financial authority. Brazil therefore offers a path toward AI capacity before the broader legal framework is settled. Build on existing regional frameworks. Echoing the Brazilian case, the Organization of American States launched the Inter-American Framework on Data Governance and Artificial Intelligence (MIGDIA), a regional initiative developed by multisector stakeholders and governments that offers guidance on data governance AI frameworks that countries can adapt even while AI law is still being drafted. One recommendation is to clearly define the processes, roles, and responsibilities of actors involved across the data lifecycle. Brazil has already taken steps in this direction by assigning a role of AI regulator to the ANPD. Officials elsewhere might consider employing MIGDIA’s framework to strengthen institutions before comprehensive AI regulation is put in place. Courts can be interim regulators. While the ANPD could become the default AI regulator under the aforementioned proposal, Brazil’s Superior Electoral Court has already played a major role as a regulatory institution during the transition. Ahead of the 2026 Brazilian elections, the court issued fourteen resolutions that governed the electoral process, including restrictions on AI use. These barred AI systems from recommending candidates or signaling voting preferences and required the disclosure of AI-generated or digitally manipulated campaign content. The court was able to enforce the resolutions through the suspension of profiles. Countries without designated AI regulators could build on this example by empowering existing judicial institutions to take on a regulatory role addressing specific risks in their wheelhouse during the transition toward an independent regulator. Develop context-specific solutions. Brazil’s first AI proposal was criticized for its insufficient inclusion of Indigenous voices in the legislative debate and failure to consider structural racism in public security. Similarly, São Paulo’s Smart Sampa city surveillance pilot project was heavily scrutinized by civil society since its presentation in 2022, as it was found to disproportionately profile black individuals. Brazil’s legislative debates incorporated concerns about racial discrimination and other contextual issues. These concerns illustrate the importance of accounting for local conditions when developing AI regulation, particularly as other Latin American countries are also piloting AI technology for security purposes. They show why institutions need clear channels for civil society participation, oversight, and accountability. The flexibility to adapt to local conditions and listen to cross-sectoral specialists can strengthen regulatory institutions over time. Lessons from Peru: institutional rollout Peru is further along than Brazil in establishing its regulatory framework. Legislation passed in 2023 establishes principles including data privacy, the development of national AI talent, and cooperation on AI development at home and abroad. The law also designates regulatory authority to the existing Secretariat of Government and Digital Transformation (SGDT). In 2025, a supreme decree approved regulations implementing the law, setting out financing provisions, accountability mechanisms, and implementation periods. The decree also clarifies how responsibilities are divided among government institutions and enforcement authorities. For example, when the SGDT encounters a regulatory breach, the decree provides the institution the authority to consult the competent authorities. Peru therefore presents a different institutional challenge from Brazil: ensuring institutions have operational capacity to enforce regulations once they are in place. Establish clear accountability measures across the regulatory ecosystem. Peru has opted to delegate implementation of its AI framework to the Secretariat of Government and Digital Transformation, with other public institutions responsible for enforcement under its regulatory umbrella. Countries adopting this approach need efficient communications channels among institutions to ensure accountability and fluidity. In Peru, different authorities manage specific infractions depending on the sector involved. For example, the National Data Protection Agency handles AI-related personal data breaches, the National Institute for the Defense of Competition and Protection of Intellectual Property handles industry competitiveness and consumer protection, and the National Police specializes in cybercrimes. Scale up support according to bandwidth. In an interview, Dilmar Villena, director of Hiperderecho, one of Peru’s leading digital safety organizations, warned the authors that without additional resources, institutions will have limited bandwidth to oversee all emerging AI technology concerns. They risk turning into hollow governing bodies that promote norms without enforcement mechanisms. Villena noted that, currently, oversight of most emerging technologies, including cloud computing and platform regulation, falls under the Secretariat of Government and Digital Transformation, which faces concerns around its budgetary capacity due to its reliance on funding from the Presidency of the Council of Ministers (PCM). Without scaling up institutional resources as regulatory responsibilities expand, AI institutions across the region risk being unable to fulfill their mandates. Maintain institutional transparency through built-in reporting. Peru has already begun doing this with the launch of its digital government indicators. These track the implementation of digital government initiatives and institutional capabilities and show where administrative capacity is thin. They do not yet measure whether AI implementation itself is working, but are an important building block for future institutional readiness assessments. Governments can use these measures to improve transparency, enhance intragovernmental cooperation, and strengthen institutional accountability. The many roads leading to AI governance Brazil and Peru offer a window into the diverse paths Latin American countries may choose for the future of AI governance. Brazil is developing authority before its regulatory framework is concretized into law. Peru has a more clearly assigned authority now that AI legislation has been approved and is beginning to be applied. The two cases point toward the same lesson: regulation is only one piece of the puzzle. Institutional readiness must be developed in tandem to avoid hollow governance. For other Latin American countries, the task is to identify where authority, operational capacity, or accountability is lagging. From there, countries that begin to close these gaps with context-specific solutions that incorporate not just government voices, but also those of civil society and technologists will be better equipped to implement responsible AI practices. Olivia Torres is a project assistant at the Atlantic Council’s GeoTech Center. Angel Ortega is a Panamanian attorney and policy researcher focused on security and defense policy, emerging technologies, AI governance, and public affairs. The GeoTech Center champions positive paths forward that societies can pursue to ensure new technologies and data empower people, prosperity, and peace. further reading Image: Map of the Americas, illustration. (Science Photo Library via Reuters Connect)

Original Source

Read the full article at Atlanticcouncil →

KhanList aggregates and links to publicly available news content. We do not host full articles from third-party sources. Always verify important information with original sources.