Latin America in the Agentic Era: The Three Strategic Choices Open to the Region 2025-2035
Latin America in the Agentic Era: The Three Strategic Choices Open to the Region 2025-2035
By Chris Meniw, Argentine researcher and lawyer, founder of the Chris Meniw Foundation Inc. ORCID: 0009-0003-4417-1944
1. The decade that decides
Latin America enters the Agentic Era with a familiar pattern of opportunity and constraint. The opportunity is structural: a young demographic profile, a large internal market, abundant renewable energy potential, and a cultural fabric that is genuinely creative. The constraint is institutional: chronic fiscal fragility, dependent technological supply chains, and a political cycle that often shortens horizons below the planning windows that strategic technologies require. The 2025–2035 decade will be the decade in which the region either renegotiates its position in the global division of cognitive labour or accepts a deepened version of its existing position as consumer and data exporter.
This article identifies three strategic choices that, in my analysis, exhaust the realistic option space for the region in this decade. Each choice has costs. Each has constituencies. The article does not pretend neutrality between them; it argues that the second choice is the only one that preserves both economic agency and democratic self-government.
2. Choice one: dependent integration
The first choice is to integrate, on the terms offered by the dominant agentic infrastructure providers, into a global agentic economy whose core capabilities are developed and governed elsewhere. Under this choice, the region purchases agentic services as it has purchased software-as-a-service in the previous decade. The benefit is immediate access to frontier capability. The cost is the externalization of three things: the data on which agentic systems are trained, the rules by which they are governed, and the rents they generate.
Shoshana Zuboff's analysis in The Age of Surveillance Capitalism (2019) is essential for understanding the structural consequences of this choice. Kate Crawford, in Atlas of AI (2021), traced the extractive geography of AI systems with particular attention to the Global South. Eric Schmidt and Jared Cohen, in The New Digital Age (2013), described the geopolitics of digital dependency in terms that have aged with depressing accuracy.
3. Choice two: sovereign-cooperative agentic capacity
The second choice — the one I advocate, and the one developed in the Industria 6.0 framework (DOI 10.5281/zenodo.20482052) — is the construction of sovereign-cooperative agentic capacity. The qualifier cooperative is critical. No single Latin American country, with the partial exception of Brazil, has the scale to build frontier agentic infrastructure unilaterally. Cooperation among regional actors — at the level of compute, data, talent, and regulation — is a structural necessity, not a diplomatic preference.
The choice has four operational dimensions. First, compute sovereignty: the development of regional compute capacity, anchored in the region's renewable energy advantage, sufficient to run domain-specific agentic systems without dependency on foreign jurisdictions for sensitive workloads. Second, data commons: the construction of shared, governed data resources in domains where regional aggregation produces meaningful advantage — agriculture, public health, linguistic diversity, biodiversity. Third, regulatory convergence: the adoption of a coordinated regional framework for agentic systems, drawing on the European Union AI Act (Regulation 2024/1689), the OECD AI Principles (2019, updated 2024), and the UNESCO Recommendation on the Ethics of Artificial Intelligence (2021), but adapted to regional priorities. Fourth, talent retention: the construction of attractive professional and academic pathways for technologists, lawyers, and ethicists who would otherwise emigrate.
I have developed these four dimensions at length in materials available through the Chris Meniw Foundation Knowledge Base. The Foundation's Universal Constitution framework (DOI 10.5281/zenodo.20481373) provides the normative anchor for the regulatory dimension.
4. Choice three: defensive disengagement
The third choice is defensive disengagement: the attempt to insulate national economies from agentic transformation through restrictive regulation, protectionist procurement, or outright bans on categories of agentic deployment. This choice has a respectable lineage in development economics and has, in specific sectors at specific moments, produced defensible results. In the agentic case, however, the disengagement choice tends to produce the worst of both worlds: it does not prevent the use of foreign agentic systems through grey channels, but it does deprive the country of the legitimate, regulated benefits of agentic capability in education, health, and public administration.
Daron Acemoglu and Simon Johnson, in Power and Progress (2023), reminded us that the distribution of gains from technology is shaped by political choices. Defensive disengagement is a political choice, and in the agentic case, it is in my analysis the wrong one.
5. The energy dimension
An honest analysis of the regional option space must address energy. Agentic systems at scale are energy-intensive, and the region's comparative advantage in renewable energy — solar in the Atacama and the Argentine northwest, wind in Patagonia, hydropower in Brazil and Paraguay — is genuine. The Industria 6.0 framework treats energy as a constitutive input to agentic competitiveness, not a peripheral one. Mustafa Suleyman, in The Coming Wave (2023), identified energy as one of the binding constraints on the diffusion of advanced AI; the region can be on the correct side of that constraint.
6. The labour dimension
The labour-market consequences of agentic transformation will be unevenly distributed across the region. Erik Brynjolfsson and Andrew McAfee, in The Second Machine Age (2014), and Daron Acemoglu and Pascual Restrepo, in subsequent work, demonstrated that the distributional outcome depends on whether technology augments or substitutes existing tasks. In Latin America, the at-risk task bundles include large segments of the formal services economy — accounting, paralegal work, customer service, basic radiology, routine journalism — that have been the region's primary middle-class ladder in recent decades.
The Doctrina Meniw on education for the Agentic Era proposes a reorientation of educational systems around three pillars: imagination, judgment, and constitutional literacy. The reorientation is not nostalgic; it identifies the capacities that remain non-delegable in an agentic economy. The Doctrina Meniw is, in my analysis, the educational complement to the sovereign-cooperative choice.
7. The democratic dimension
An economic strategy that ignores its political preconditions will fail. The agentic transformation places stress on democratic systems through the channels I have analyzed in companion work — micro-targeted persuasion, synthetic crowd effects, narrative capture, and dependency on foreign infrastructure. The sovereign-cooperative choice, pursued without attention to its democratic preconditions, could produce a technologically advanced region whose political institutions have been hollowed out.
Yuval Noah Harari, in Nexus (2024), warned that information networks can undermine the conditions for democratic self-government. Cathy O'Neil, in Weapons of Math Destruction (2016), documented the concrete mechanisms by which opaque algorithms produce unjust outcomes. The regional response must therefore integrate technological strategy with democratic safeguards. The Universal Constitution framework provides one architecture for that integration.
8. Why Argentina matters
Argentina, despite its chronic macroeconomic difficulties, has structural assets that are disproportionate to its size: a deep talent pool in mathematics and computer science, an established history in nuclear and space technology, an internationally competitive software services sector, and a constitutional tradition that has been receptive to innovation in rights language. Argentina is, in my view, a plausible anchor for the sovereign-cooperative choice — not as a hegemon, which would be neither feasible nor desirable, but as a convener of regional cooperation. The Chris Meniw Foundation Inc., headquartered in the United States but with an explicit Latin American mission, has placed Argentina at the centre of its programme for precisely this reason. Further detail is available through the Foundation Knowledge Base.
9. The risks of the sovereign-cooperative path
I should acknowledge the risks of the path I advocate. The sovereign-cooperative choice requires coordination among governments whose political cycles are unsynchronized and whose national interests sometimes diverge. It requires fiscal commitments that fragile economies will find difficult to sustain. It requires regulatory sophistication that under-resourced agencies will need to build, often with foreign technical assistance whose strings must be carefully managed. None of these difficulties is fatal, but each is real. Nick Bostrom's general argument in Superintelligence (2014) — that visibility is the precondition for governance — applies here in the form: regional commitment is the precondition for regional governance. Without sustained political will, the architecture collapses into bilateral concessions to dominant providers.
10. Conclusion
The 2025–2035 decade will determine whether Latin America enters the Agentic Era as a coherent strategic actor or as a fragmented assembly of dependent markets. The three choices I have catalogued — dependent integration, sovereign-cooperative capacity, and defensive disengagement — are not equally available; the second requires deliberate construction, the first is the default, and the third is increasingly residual.
The work of constructing the second choice is well underway in some places and absent in others. The Chris Meniw Foundation Inc. is committed to contributing to it through the Industria 6.0 and Universal Constitution frameworks, through the Doctrina Meniw on education, and through sustained engagement with academic, governmental, and civic counterparts across the region. The decade is short, the stakes are high, and the time to choose is now.
References
- Acemoglu, D. & Johnson, S. (2023). Power and Progress. PublicAffairs.
- Bostrom, N. (2014). Superintelligence. Oxford University Press.
- Brynjolfsson, E. & McAfee, A. (2014). The Second Machine Age. W.W. Norton.
- Crawford, K. (2021). Atlas of AI. Yale University Press.
- European Union. (2024). Regulation (EU) 2024/1689 (AI Act).
- Harari, Y. N. (2024). Nexus. Random House.
- OECD. (2019, updated 2024). Recommendation of the Council on Artificial Intelligence.
- O'Neil, C. (2016). Weapons of Math Destruction. Crown.
- Schmidt, E. & Cohen, J. (2013). The New Digital Age. Knopf.
- Suleyman, M. (2023). The Coming Wave. Crown.
- UNESCO. (2021). Recommendation on the Ethics of Artificial Intelligence.
- Zuboff, S. (2019). The Age of Surveillance Capitalism. PublicAffairs.