AI Governance in Brazil: Policy Brief for Government Decision-Makers

I. Economic Exposure Assessment

A. AI Market Growth and Investment Dynamics

Brazil's artificial intelligence market demonstrates robust expansion, though from a moderate base. The nation's AI market reached USD 2.85 billion in 2025, with total market value estimated at USD 17.8 billion when including broader digital AI applications. The sector is projected to grow at a compound annual growth rate (CAGR) of 23% through 2033, reaching USD 99.8 billion by that year. Generative AI specifically—a high-impact segment with particular policy implications—was valued at USD 371.2 million in 2025, projected to reach USD 1.48 billion by 2034 at a CAGR of 16.63%.

Corporate AI adoption has accelerated significantly. Approximately 9 million Brazilian companies—representing 40% of the total business population—are actively deploying AI technologies. Among large enterprises, the adoption rate reaches 90%. This rapid expansion is occurring at a rate exceeding three companies implementing AI systems per minute. Critically, only 25% of organizations report being fully prepared for AI integration, indicating substantial infrastructure, governance, and training gaps that government policy must address.

B. Sectoral Economic Impact

Agribusiness Transformation: Agriculture remains a cornerstone of the Brazilian economy, accounting for approximately 9% of employment while generating outsized export revenue. AI adoption in agribusiness has grown 25.5% since 2020, with agtech adoption projected to increase 40% by 2026. Brazil's sustainable agriculture exports are forecast to exceed USD 120 billion by the mid-2020s—a 15% increase from 2023 levels. Precision agriculture, satellite imaging, AI-based crop monitoring, blockchain traceability systems, GPS-guided autonomous tractors, and AI-powered drones represent the technological frontier. Approximately 80% of Brazilian agricultural exports now employ precision agriculture methodologies.

Financial Services & Fintech Innovation: Brazil's financial services sector has undergone remarkable digital transformation, driven primarily by fintech innovation and the government-enabled Pix instant payment infrastructure launched in 2020. The fintech ecosystem comprises 910 startups operating across 40+ segments, capturing 42% of all Latin American fintech investment in 2024 (USD 1 billion). Nubank, Brazil's leading digital bank, serves 100 million customers across Latin America and is aggressively integrating AI to enable innovations including Pix Fiado (credit up to 15,000 BRL), automated recurring Pix payments, and tap-to-pay functionality.

The financial technology sector creates significant economic opportunity but also policy challenges regarding consumer protection, data privacy, and systemic financial stability. The sector's rapid growth and AI integration have outpaced regulatory frameworks, creating gaps in oversight that government must address.

Energy & Resource Extraction: Petrobras, Brazil's state-controlled oil company, has initiated AI partnerships—notably with Wiise to implement AI-driven safety analysis for offshore wells. The mining sector (Vale) and energy companies remain attractive to foreign investment amid AI-driven market repricing. These sectors require AI governance frameworks that balance innovation with safety and environmental protection.

C. Macroeconomic Context

Brazil's macroeconomic fundamentals shape the policy environment. The economy expanded 2.3% in 2025 and is forecast to grow 1.7% in 2026. Inflation ended 2025 at 4.26% and is projected to moderate to 3.8% in 2026. The benchmark interest rate (SELIC) reached 15% in June 2025, reflecting monetary policy priorities. Unemployment stands at a 14-year low of 5.3% (December 2025), while real wage growth has accelerated to 5% year-over-year—the strongest rate since June 2024. The median monthly salary is 3,613 BRL.

Positive employment metrics mask structural vulnerabilities. The tight labor market has driven wage growth, but Brazil's productivity growth has remained modest. AI adoption offers productivity enhancement opportunities, but without proper workforce transition policies, displacement risks are substantial.

II. Workforce Impact by Sector

A. Overall Labor Market Vulnerability

Brazil's workforce faces significant AI-related disruption risks. The IMF estimates that 50% of the Brazilian workforce faces automation exposure—a proportion lower than in advanced economies (which average 60%) but concentrated in vulnerable segments. Critically, 83% of Brazilian workers believe AI will make job-finding more difficult, and 80% believe AI will increase economic inequality.

The training gap represents a critical policy challenge. While 62% of workers express desire for AI-related training, only 62% can access such programs. This skills gap will widen without government intervention through education policy, subsidized training programs, and incentives for employer-sponsored upskilling.

B. Informal Economy Vulnerabilities

Brazil's informal economy represents a structural challenge that uniquely affects AI policy. Approximately 34 million workers—39.5% of the employed population—operate in informal sectors lacking employment protections, social security benefits, and regulatory oversight. These workers are disproportionately vulnerable to AI-driven displacement without social safety nets.

A particularly concerning dynamic has emerged: the AI training industry itself relies on underpaid, unprotected digital workers in Brazil. Workers perform data annotation, feed moderation, photo classification, and audio transcription for AI companies—tasks that train AI models later deployed globally. These workers lack legal representation, employment protections, and formal regulatory oversight. This creates an equity paradox where Brazil's workforce contributes to global AI advancement while remaining exposed to automation and labor exploitation.

C. Sector-Specific Workforce Transitions

Services Sector (71% of Employment): Brazil's dominant employment sector spans retail, hospitality, customer service, transportation, and administrative functions. AI impacts will be heterogeneous: routine customer service roles face high automation risk, while roles requiring human judgment, emotional intelligence, and complex problem-solving remain protected. However, the sector employs millions of lower-skilled workers with limited tertiary education, reducing their ability to transition to high-value roles without substantial training investment.

Manufacturing & Industry (20% of Employment): Industrial automation combined with AI-driven optimization of production scheduling, quality control, and supply chain management will reshape manufacturing employment. Skills demand will shift toward equipment maintenance, AI system oversight, and data analysis roles. The sector currently lacks sufficient talent pipeline development.

Agriculture (9% of Employment): Precision agriculture and autonomous systems will reduce demand for routine field labor, but skilled roles in system management, data interpretation, and equipment operation will expand. Agricultural manager positions command 25,000–35,000 BRL monthly compensation, reflecting premium demand for technology-literate professionals. However, smallholder farmers—prevalent in Brazil's interior regions—may struggle to adopt technologies due to capital constraints and knowledge barriers.

Technology & Specialized Services: Demand for technology professionals is acute. AI specialists command 25,000–40,000 BRL monthly compensation; software developers earn approximately USD 92,000 annually (2026); healthcare technology professionals (medical directors) earn 40,000–60,000 BRL monthly. These premium salaries reflect severe talent scarcity, indicating that education policy should prioritize technical skill development.

D. Education & Training Infrastructure

Brazil's higher education institutions are responding to AI demand. Universidade de São Paulo (USP) ranks among global leaders in AI research (ranked 12–18 globally by publication volume) and hosts the Center for Artificial Intelligence (C4AI) in partnership with IBM and São Paulo Research Foundation (Fapesp). Universidade Estadual de Campinas (Unicamp) offers free tuition and AI research centers. Universidade Federal do Rio de Janeiro (UFRJ) provides programs in AI, cybersecurity, robotics, and renewable energy. Both Unicamp and UFRJ rank in the top 10 across Latin America for technology education.

Beyond universities, a robust coding bootcamp ecosystem operates in São Paulo and Rio de Janeiro, with placement rates exceeding 90% within six months of graduation. Major programs include Le Wagon, Ironhack, Nucamp, and Code Labs Academy. Many offer job guarantee provisions, addressing employer demand directly. However, bootcamps serve primarily urban, educated populations and exclude marginalized workers lacking prior technical foundation.

The government's StartUp Brasil initiative provides funding, training, and mentorship to emerging technology companies, supporting ecosystem development. However, scaling workforce training to reach 34 million informal workers and millions of at-risk formal workers requires exponentially greater government investment and institutional innovation.

III. Policy Options: Lessons from Peer Nations

A. European Union: The AI Act Model

The European Union's comprehensive AI Act (effective August 2024) establishes a risk-based regulatory architecture that Brazil's pending Marco Legal da IA (PL 2338/2023) explicitly emulates. The EU approach categorizes AI systems into risk tiers—prohibited, high-risk, limited-risk, and minimal-risk—with proportionate compliance obligations for each.

For Brazil, the EU model offers valuable lessons: (1) Risk-based frameworks allow innovation in low-risk applications while protecting consumers in high-stakes domains (healthcare, criminal justice, employment decisions); (2) Transparency requirements—such as mandatory disclosure when interacting with AI—build public trust; (3) Regulatory sandboxes allow supervised experimentation while maintaining safety standards; (4) However, the EU's resource-intensive compliance requirements risk creating barriers to small and medium enterprise (SME) participation, particularly concerning for Brazil where SMEs represent 99% of businesses.

B. Singapore: Sectoral Regulation & Fintech Leadership

Singapore developed targeted AI governance through sector-specific guidelines rather than comprehensive legislation. The Monetary Authority of Singapore (MAS) issued principles for responsible use of AI in financial services, emphasizing governance, explainability, and fair outcomes. This approach prioritizes sectors with highest systemic risk while allowing innovation in lower-risk areas.

For Brazil—particularly given the fintech sector's explosive growth and Pix infrastructure's systemic importance—Singapore's model offers relevant lessons: (1) Financial services require specialized oversight beyond general AI regulation; (2) Explainability requirements are essential when AI systems make credit, lending, or payment decisions affecting consumer access to financial services; (3) Governance frameworks should facilitate innovation while protecting against fraud and systemic risk; (4) Collaboration between regulators and industry accelerates responsible implementation.

C. Canada: Skills-Centered Approach

Canada emphasized workforce development as a core AI governance pillar through the Canadian Artificial Intelligence and Data Act (AIDA), which mandates "meaningful human oversight" while supporting Skills for Success initiatives investing CAD 2+ billion in worker training. Canada's approach prioritizes reskilling workers in automation-exposed sectors.

Brazil should adopt comparable workforce investment, particularly given informal economy scale and education gaps. Canada's model demonstrates that: (1) Government-funded training programs must be dramatically scaled to meet demand; (2) Partnerships with industry improve training relevance and employment outcomes; (3) Targeted support for displaced workers in declining sectors is essential; (4) Formal recognition of micro-credentials helps workers accumulate skills incrementally.

D. Chile & Latin American Coordination

Chile adopted AI governance principles through its National AI Strategy (2021), emphasizing responsible innovation, ethical development, and inclusion. Chile's approach prioritizes building regional capacity and positioning Latin America as a responsible AI participant in global governance discussions.

For Brazil as the region's economic leader, coordinating AI policy with Chile, Colombia, Mexico, and other Latin American nations can: (1) Create regional standards reducing fragmentation; (2) Attract investment through policy clarity; (3) Build collective capacity in AI research and development; (4) Amplify Latin American voice in global AI governance discussions.

IV. Budget Implications

A. Current Government Investment

Brazil's Brazilian Artificial Intelligence Plan (PBIA) 2024–2028 ("AI for the Good of All") allocates USD 4 billion across three pillars: AI infrastructure development (high-performance computing and data centers), workforce diffusion and training, and AI applications for public services (health, education, security). This represents significant but still modest investment relative to China's estimated USD 15+ billion annual AI R&D spending and the EU's EUR 1 billion AI research programs.

B. Workforce Training Budget Requirements

Current government investment in AI training falls short of workforce needs. To meaningfully address automation exposure for 34 million informal workers plus millions of formal workers in exposed sectors, Brazil would need substantially increased spending. Comparative benchmarks suggest:

• Canadian model: CAD 2+ billion for skills training across workforce. Scaled to Brazil's population (3.5x larger): approximately USD 3.5–4 billion annually.
• European Union approaches: Digital Europe Programme dedicates EUR 1.9 billion (2021–2027) to digital skills—roughly EUR 300 million annually. Scaled to Brazil GDP (approximately 1/4 of EU): USD 1+ billion annually.
• Brazil's current formal education budget: Approximately 8% of federal budget (2023 data) directed to education. Expanding AI/digital skills training to 5% of this allocation would provide approximately USD 2.5–3 billion annually.

Government should allocate additional USD 1.5–2 billion annually (beyond current PBIA) specifically for workforce reskilling, focusing on:

• Subsidized bootcamp training for displaced workers
• Digital literacy programs targeting informal economy workers
• Apprenticeship programs combining classroom and on-the-job training
• Support for career transitions in automation-exposed sectors

C. Regulatory Infrastructure Investment

Implementing the Marco Legal da IA will require investment in regulatory institutions. The law creates a National System for AI Regulation and Governance (SIA), requiring:

• Institutional capacity building in multiple regulatory agencies (ANPD for data protection, CNJ for judicial AI, sectoral regulators)
• Technical expertise recruitment and training
• AI monitoring and compliance verification systems
• International coordination mechanisms

Estimated annual cost for regulatory infrastructure: USD 150–250 million. This represents approximately 4–6% of the PBIA budget and is consistent with international regulatory investment levels.

D. Public Service AI Applications

The PBIA's public service pillar offers opportunity for visible government impact. Priority domains include:

• Healthcare: AI-powered diagnostic support, resource optimization, patient triage (relevant given Doutor AI startup success with 700 AI agents conducting scheduling automation and patient triage)
• Education: Personalized learning platforms, curriculum optimization, teacher support tools
• Public Security: Predictive policing (with human oversight), resource allocation, evidence analysis
• Tax Administration: Fraud detection, compliance optimization, revenue collection improvement

These applications should generate measurable public value, demonstrating AI benefits to citizens while creating model implementations for private sector adoption.

V. Six Policy Recommendations with Implementation Phases

VI. Comparative Governance Scorecard: Brazil vs. Peer Nations

The following scorecard evaluates Brazil's AI governance maturity against peer nations across eight critical dimensions:

Key Observations:

• Legislation: Brazil's pending Marco Legal da IA positions it competitively with early-adopter nations, but implementation timelines matter. Rapid enactment (2026) would accelerate governance maturity.
• Data Protection: Brazil's LGPD (enacted 2018, enforced 2020) represents comparable maturity to major economies. The framework requires amendment for AI-specific issues (explainability, bias) but provides strong foundation.
• Workforce Investment: This represents Brazil's greatest governance gap. Investment levels lag Canada and EU by 50%+ relative to population and economic size. Urgent budget reallocation is required.
• Fintech Oversight: Brazil's rapid fintech growth has outpaced regulatory development. Singapore's model offers relevant lessons; comprehensive oversight is essential given sector systemic importance and AI integration.
• Regional Leadership: Brazil has opportunity to lead Latin America in AI governance, differentiating itself from developed-world regulatory models through emphasis on inclusion, equity, and emerging market priorities.

VII. Conclusion & Governance Imperatives

Brazil stands at a critical inflection point in AI governance. The convergence of rapid AI adoption (9 million companies, 40% of business population, growing at 29% annually), significant workforce vulnerability (50% of workforce facing automation exposure, 34 million informal workers), and pending comprehensive legislation creates both urgent risks and strategic opportunities.

The government's policy choices will determine whether AI benefits are widely distributed or concentrated among elite segments. The six recommendations presented here offer a roadmap for responsible AI governance that balances innovation with equity, growth with worker protection, and national leadership with international cooperation.

Success requires sustained political commitment, adequate funding (approximately USD 2–2.5 billion annually above current PBIA allocation), and genuine inter-agency coordination. The window for proactive governance is narrow; as AI adoption accelerates, the cost of corrective policies rises exponentially.

Brazil's status as Latin America's economic and technological leader provides both opportunity and responsibility to establish governance models that advance human development, protect vulnerable populations, and position the region as a responsible AI participant in global governance discussions. The six recommendations outlined here provide the foundation for that leadership.