This article is part of The AI Revolution in Latin America, a series that addresses what steps Latin America needs to take in order to effectively implement AI and further digitalize the region.
Artificial intelligence presents Latin America with a historic opportunity to accelerate growth, attract investment, modernize governments, and close persistent development gaps. But realizing that potential will depend as much on institutions as on technology. Rather than importing regulatory models from the United States, Europe, or China, the region should build a predictable, risk-based, and interoperable framework grounded in the rule of law. Clear rules, strong institutions, regulatory capacity, and regional coordination can simultaneously protect citizens, foster innovation, and turn legal certainty into a competitive advantage for attracting AI investment.
Artificial intelligence is often framed as a technological race. Who will build the largest models, attract the most data centers, or deploy AI fastest across government and business?
For Latin America, however, the more consequential race may be institutional. The region’s ability to benefit from AI will depend less on copying the regulatory models of Washington, Brussels, or Beijing than on whether it can build rules that are clear, predictable, enforceable, and compatible across borders.
This is a moment of extraordinary opportunity. AI could allow emerging economies to leapfrog traditional development constraints by improving healthcare and education, increasing agricultural productivity, modernizing public services, and giving small businesses access to capabilities previously available only to large corporations.
But opportunity and disruption will arrive together. Bill Gates recently warned that AI could become either an extraordinary equalizer or a powerful source of inequality. Unlike previous technological transformations, this one could move remarkably fast because AI operates through devices people already own, uses natural language, and can be incorporated rapidly into existing economic processes. Governments therefore cannot afford to wait for disruption before designing their response.
For Latin America, that warning is particularly important. A region characterized by high informality, persistent inequality, limited fiscal space, and fragile institutions cannot approach AI simply as another technology policy. AI governance must become part of a broader development strategy.
Regulatory uncertainty also carries an economic cost. Investors considering data centers, cloud infrastructure, AI services, or research hubs do not only look at electricity prices, connectivity, talent and market size. They ask basic questions such as, which rules will apply? Who will enforce them? Can decisions be appealed? Will regulations change abruptly after an election? Will one country’s compliance regime be recognized in another?
These are fundamentally questions of the rule of law related to AI.
Good AI governance should therefore be understood not as a brake on innovation but as economic infrastructure. Eduardo Levy Yeyati has argued that smart regulation can encourage adoption and investment when it creates trust and predictable conditions. The OECD similarly emphasizes that trustworthy AI requires both “enablers,” including infrastructure, skills, investment, and institutional capacity, and “guardrails,” such as transparency, accountability, and risk management.
Latin America currently risks moving in the opposite direction, toward regulatory fragmentation. A 2026 study by Universidad de los Andes in Bogota identified 716 AI-related regulatory instruments, rules, and legislative proposals across 23 countries and territories, more than 80 percent of which were initiated since 2023. Legislative activity is welcome, but hundreds of overlapping initiatives can produce uncertainty rather than governance.
The region should therefore resist the false dichotomy of choosing between regulation and innovation. Poor regulation can discourage investment, but so can regulatory ambiguity, arbitrary enforcement, and weak institutions. Nor should Latin America simply choose between the European Union’s rights-centered model and the United States’ more innovation-driven approach. It should develop its own adaptive, risk-based framework anchored in the rule of law.
Five actions or recommendations should guide that strategy:
First, regulate risks, not technologies. High-risk applications—including AI used in criminal justice, policing, healthcare, critical infrastructure or access to public benefits—require stronger transparency, human oversight, and mechanisms for appeal. Low-risk commercial applications should face lighter requirements.
Second, build institutions before regulatory complexity. Gates correctly argues that existing bureaucracies were not designed to manage a technology that simultaneously affects employment, taxation, education, national security, energy, and financial systems. Latin American governments need high-level AI governance mechanisms capable of coordinating ministries, regulators, and the private sector. Passing sophisticated legislation without technical expertise and enforcement capacity will create uncertainty rather than protection.
Third, create regulatory sandboxes with legal certainty. Companies, universities and startups should be able to test AI products under supervised conditions without fearing retroactive or contradictory regulation. Innovation and accountability can coexist when the rules are known in advance.
Fourth, pursue regional interoperability. Latin America does not necessarily need a single AI law, but it does need common definitions, risk categories, audit principles, and mutual recognition mechanisms. Regulatory compatibility would lower compliance costs and transform a collection of relatively small national markets into an attractive AI ecosystem of more than 600 million people.
Fifth, make inclusion part of AI governance from the beginning. Gates’ central warning deserves particular attention in Latin America. The benefits of AI will not distribute themselves automatically. Governments should anticipate labor displacement, invest aggressively in reskilling, expand digital infrastructure, support AI tools in Spanish, Portuguese, and indigenous languages, and prioritize applications that improve education, healthcare, agriculture, and public services. AI policy should be measured not only by investment attracted or startups created, but also by whether productivity gains reach citizens who have historically been excluded from technological progress.
Steve Jobs understood technological inflection points unusually well. In a remarkable 1983 speech revisited by Steven Levy in WIRED, Jobs anticipated that computers would become deeply embedded in everyday life long before most people understood their significance. AI represents an even larger turning point.
Latin America should not spend this moment debating whether to embrace or resist it. The real question is whether the region can build institutions capable of managing its consequences. The countries that succeed will not necessarily be those with the most AI laws. They will be those that combine innovation with inclusion, predictable regulation with institutional capacity, and technological ambition with democratic accountability.
In the age of artificial intelligence, the rule of law may become one of Latin America’s most important competitive advantages.
Luis Gilberto Murillo is a Colombian diplomat, public policy leader, and former Minister of Foreign Affairs of Colombia. He previously served as Colombia’s Ambassador to the United States, Minister of Environment and Sustainable Development, and Governor of Chocó. An Afro-Colombian leader with extensive experience in diplomacy, peacebuilding, environmental policy, and sustainable development, Murillo is also a member of Global Americans’ International Advisory Council.
Image via CAF
