Volume 7: Governance and Standards

The preceding six volumes have examined how AI reshapes industries, displaces workers, demands unprecedented resources, and transforms military power. But technology doesn't develop in a vacuum. Every capability discussed in this series exists within governance frameworks that determine what's permissible, what's required, and what triggers sanctions.

Governance frameworks aren't mere bureaucratic constraints. They're instruments of power as consequential as supply chains or weaponry. The nation or bloc that sets standards shapes how everyone else develops and deploys AI. The regulatory model that becomes dominant determines which companies can compete globally and which face fragmentation into incompatible markets. The governance choices made today will create path dependencies that persist for decades.

For the Americas, governance presents both urgent challenges and strategic opportunities. The hemisphere spans democracies and authoritarian systems, advanced economies and developing nations, technology leaders and technology consumers. These differences create friction over what AI governance should accomplish, who should write the rules, and whether hemispheric harmonization is even possible.

Politics, Geography, and the Ballot Box

AI governance doesn't emerge from abstract principle. It's shaped by political structures, geographic realities, and electoral outcomes.

Federal systems like the United States, Canada, Mexico, and Brazil create multi-level governance where national laws, state or provincial rules, and municipal experiments all shape the regulatory landscape. This can drive innovation as different jurisdictions test approaches, but it also creates compliance complexity for firms operating across borders.

Geography matters because AI infrastructure concentrates in specific locations. Data center clusters and digital infrastructure routes create regulatory gravity wells. Regions that host major AI facilities gain leverage over how systems operate, while regions without infrastructure become rule-takers dependent on decisions made elsewhere.

Regional trade frameworks like USMCA and MERCOSUR create pressure for regulatory harmonization. Once one key member adopts particular AI governance standards, neighbors face strong incentives to align to preserve market access and interoperability.

Elections decide not only who governs but what kind of information ecosystem citizens will inhabit for decades. Winners set rules on algorithmic transparency, content moderation, digital identity, and surveillance that embed values into technical systems difficult to change later. In the United States, shifts between administrations drive dramatically different emphases. The Trump administration's January 2025 executive order "Removing Barriers to American Leadership in Artificial Intelligence" and December 2025's "Ensuring a National Policy Framework for Artificial Intelligence" prioritize innovation and federal preemption of state rules. A future administration could reverse course entirely, imposing comprehensive federal AI regulation that currently doesn't exist.

In Mexico, constitutional reform proposals would explicitly add AI to Congress's legislative remit, meaning a single legislative majority could restructure the governance architecture for decades. AI-driven campaigning itself forces governance responses. Micro-targeting, deepfakes, and algorithmic manipulation create pressure on electoral authorities to develop AI standards for campaign technology, shaping democratic resilience for future generations.

Canada: Standards-Led Rights Model

Canada has built its AI governance on strong privacy and human rights foundations, positioning itself as a bridge between American innovation priorities and European precautionary frameworks.

The Artificial Intelligence and Data Act (AIDA), originally introduced as Part III of Bill C-27 in 2022, died when Parliament was prorogued on January 6, 2025, ahead of federal elections. The new government under Prime Minister Mark Carney appointed Evan Solomon in May 2025 as the first minister responsible for AI and Digital Innovation, signaling continued focus even as the legislative pathway requires restarting.

Provincial action has driven practice even without federal AI law. Québec's Law 25 imposes algorithmic accountability and automated processing requirements that influence national standards. When provinces hosting major AI research centers adopt strict rules, federal approaches must accommodate or risk internal conflicts.

In Budget 2024, Canada allocated CAD2.4 billion to advance AI, with CAD2 billion dedicated to an AI Compute Access Fund. Canada's 2025 national standards on accessible and equitable AI push for transparency, bias mitigation, and inclusion. These aren't merely technical documents but soft power instruments that export a template for human-centric AI compatible with EU-style regimes, positioning Canada as a bridge between U.S. and European approaches.

The September 2023 Voluntary Code of Conduct on the Responsible Development and Management of Advanced Generative AI Systems had attracted more than 55 organizations by mid-2025. But voluntary commitments represent stopgaps, not comprehensive governance, creating uncertainty about what binding requirements will eventually emerge.

Canada's fundamental challenge is sovereignty. Can a country of 40 million people, economically integrated with a neighbor pursuing minimal regulation, maintain independent governance standards that meaningfully constrain AI development? Or will market forces render Canadian rules largely symbolic?

United States: Innovation-First Plus Enforcement

The United States pursues AI governance through sectoral enforcement, voluntary frameworks, and increasingly aggressive federal preemption of state action.

Rather than comprehensive AI legislation, the U.S. system relies on sectoral regulators using existing authorities. The Federal Trade Commission polices consumer protection. The Consumer Financial Protection Bureau addresses credit and lending. The Equal Employment Opportunity Commission handles workplace discrimination. Each agency brings its own statutory authorities to bear on AI applications within its domain.

The National Institute of Standards and Technology's AI Risk Management Framework provides voluntary but influential guidance that many companies adopt as de facto standards. State-level legislation has proliferated, with more than 1,000 AI-related bills introduced across all U.S. states and territories in 2025.

The December 2025 Executive Order 14365 "Ensuring a National Policy Framework for Artificial Intelligence" represents a dramatic shift toward federal preemption. It establishes federal policy "to sustain and enhance the United States' global AI dominance through a minimally burdensome national policy framework."

The order directs the Attorney General to establish an AI Litigation Task Force to challenge state AI laws deemed inconsistent with federal policy. It instructs the Department of Commerce to evaluate existing state laws and identify those deemed "onerous" for potential legal challenge. Most controversially, it directs agencies to condition federal funding on states not enacting conflicting AI laws or agreeing not to enforce such laws during grant performance periods.

The U.S. approach reflects powerful political economy factors. Strong technology sector lobbying and geopolitical competition with China push Washington toward lighter regulatory touch compared with the EU. The emphasis falls on security, innovation, and voluntary commitments rather than comprehensive ex-ante regulation. This model prioritizes dynamic innovation and post-hoc enforcement, betting that faster commercial experimentation outweighs risks of uneven protection.

Mexico: Sovereignty and Institution-Building

Mexico occupies a distinctive position, caught between North American economic integration and desires for technological sovereignty.

Mexico has seen many AI-related bills introduced since 2020 but lacks unified AI law. The most significant development is the 2025 proposal to amend the Constitution to explicitly empower Congress to legislate on AI and enable a General Law on AI. This constitutional approach reflects Mexican legal tradition where framework legislation requires explicit constitutional authorization.

The proposed amendment's objectives emphasize responsible innovation, human rights protection, national security, and technological sovereignty. These signal intent to use AI governance as a development tool while asserting control over technological trajectory rather than simply importing foreign frameworks.

As a USMCA member, Mexico must balance alignment with North American requirements against domestic pressures for sovereignty. The December 2025 U.S. executive order compounds this challenge. If Washington conditions security cooperation or technology transfers on alignment with American AI governance principles that emphasize minimal regulation, Mexico faces difficult choices between accommodating U.S. preferences and responding to domestic constituencies demanding protections from AI harms.

Brazil: Regional Rule-Maker

Brazil has emerged as Latin America's leader in AI governance, driven by its size, economic weight, and ambition to project itself as a standard-setter for the Global South.

The Brazilian AI Act, Bill No. 2338/2023, approved by the Senate in December 2024 and forwarded to the Chamber of Deputies, represents one of the world's most comprehensive AI governance frameworks outside Europe. The bill adopts a risk-based approach classifying AI systems into minimal, limited, high, and unacceptable risk categories.

Key provisions include transparency and explainability requirements for high-risk systems, user rights to contest automated decisions, and obligations on AI agents across the lifecycle. The framework aims to harmonize with Brazil's General Data Protection Law (LGPD). The National Data Protection Authority (ANPD) is expected to play a central role in enforcement.

The Brazilian Artificial Intelligence Strategy (EBIA), approved in 2024, allocates approximately BRL23 billion (around USD4 billion) over four years to support infrastructure development, capacity building, business innovation, and regulatory improvement.

Brazil's approach emphasizes developmental uses of AI in agriculture, fintech, public services, and infrastructure. Regulatory sandboxes in sectors like finance and health allow controlled experimentation. Brazil seeks to position itself as a standard-setter for the Global South, aligning partly with EU principles but emphasizing developmental flexibility and sensitivity to inequality and institutional capacity constraints.

Argentina, Chile, and Emerging Players

Mid-sized economies across Latin America are developing governance approaches, often borrowing from multiple models while adapting to local circumstances.

Argentina remains in a proposal-heavy stage, drawing inspiration from EU-like risk models and human rights language while not yet having comprehensive AI legislation in force. Political and economic volatility complicate governance development, creating uncertainty about whether proposed frameworks will survive implementation.

Chile functions as a bellwether for the Southern Cone due to relatively advanced digital governance and strong data protection agenda. Chile's political stability and extensive trade agreements position it to test governance approaches that bridge different regulatory philosophies.

Colombia and Uruguay have developed emerging AI strategies and experiments with sandboxes and sector-specific guidelines. These pragmatic approaches borrow from EU and OECD playbooks while keeping investment attractiveness in mind.

These countries illustrate how regional convergence might happen not through a single treaty but through a mesh of similar principles. Risk-based regulation, human rights framing, transparency obligations, and sandbox experimentation appear across multiple countries, suggesting potential for harmonization even without formal coordination.

Three Competing Models

The diverse approaches across the Americas reflect three competing logics. Canada pursues a standards-led rights model emphasizing human rights, privacy, and inclusion, using comprehensive privacy rules, national AI standards, and strong regulators as implementation tools. The United States embraces innovation-first enforcement that maximizes competitiveness through agency enforcement, voluntary frameworks, and federal preemption rather than ex-ante licensing. Mexico, Brazil, Argentina, and Chile experiment with sovereignty-and-development models that use AI governance to climb value chains while asserting data and technological sovereignty through constitutional reforms, risk-based bills, and sandboxes.

These models compete for normative influence. Multinational firms must choose which standards to adopt as defaults. Trade agreements and technical assistance programs spread particular templates. Countries with more successful governance frameworks attract investment and talent while those with dysfunctional approaches face capital flight.

Why Divergence Matters

The abstract debates over regulatory philosophy take on concrete importance when neighboring countries adopt incompatible AI governance frameworks.

Divergent AI rules raise compliance costs and legal uncertainty for firms operating across borders, particularly in integrated supply chains like USMCA. A company managing operations across U.S., Canadian, and Mexican facilities must satisfy different transparency requirements, different standards for algorithmic accountability, and different liability regimes depending on where decisions occur.

Differences in data protection and algorithmic governance can fragment digital markets, encourage inefficient data localization, and complicate cross-border AI-enabled services. Workers, entrepreneurs, and researchers may gravitate toward jurisdictions perceived as offering better digital rights or conversely toward lightly regulated hubs providing rapid growth. As AI reshapes labor markets, countries with weaker governance may see regulation-flight of both firms and talent.

Governance incompatibility can create political tensions across borders. If one country's AI surveillance systems operate under different standards than its neighbor's, migrants and asylum seekers face unpredictable treatment depending on which side of a border they're on. Within countries, inadequate AI governance that allows unchecked surveillance, algorithmic discrimination, and labor displacement without support can fuel political radicalization and democratic backsliding.

The Ongoing Nature of Governance

AI governance in 2026 remains profoundly unsettled. New technologies constantly expose gaps in existing rules. Frontier models and highly autonomous systems raise questions about liability and control that existing frameworks don't adequately address.

Standards evolve as organizations develop new management system standards and countries establish national benchmarks. Political cycles and crises trigger episodic regulatory surges. Election interference, major AI accidents, and labor market shocks create windows for governance action that might not exist during calmer periods.

Emerging economies in the Americas use AI governance strategically beyond simply protecting citizens. They bargain over data localization, cloud service conditions, and infrastructure investments with large U.S., European, and Chinese providers, using governance frameworks as leverage. They promote digital industrial policies through special regimes for AI startups and public sector procurement standards that favor local solutions.

Many Latin American proposals blend imported concepts like risk-based classification with local concerns about inequality, informality, and access gaps. The result is hybrid frameworks that reflect specific development contexts rather than fitting cleanly into either European or American models.

What Comes Next

Governance and standards represent the legal and institutional frameworks within which all the previous volumes' dynamics play out. But governance is also where values and power collide most directly. The technical questions about AI regulation are simultaneously political questions about what kind of society we're building and who gets to decide.

For the Americas, the governance challenge is whether the hemisphere can develop frameworks that enable innovation while protecting citizens, that facilitate economic integration while preserving national sovereignty, and that adapt to emerging technologies while maintaining democratic accountability. These are not easy balances to strike. The competing models reflect genuinely different priorities and constraints.

The choices being made today about AI governance will shape the hemisphere's trajectory for generations, determining which nations rise, which fall behind, and whether the Americas can maintain the democratic institutions and human rights protections that should distinguish the region from authoritarian alternatives.

Next week, we turn to perhaps the most difficult topic in this series: AI sovereignty and dependency in the Global South. We'll examine whether Latin American nations can chart independent courses in AI development or whether technological dependence is inevitable. The governance frameworks discussed today shape what's possible, but they don't determine outcomes. That depends on choices yet to be made.

This is Part 7 of a 10-part series on The Geopolitics of Artificial Intelligence in the Americas by Core Geopolitical Insights LLC. Follow along each week as we explore how this transformative technology is reshaping power, prosperity, and security across the Western Hemisphere. | Next week: AI Sovereignty and Dependency in the Global South