The Silent Retreat of AI in the Dominican Republic
A critical reading of the Dominican Republic's position in the Latin American Artificial Intelligence Index (ILIA) 2025 and what it implies for our regional competitiveness. The recent Latin American Artificial Intelligence Index (ILIA) 2025, produced by ECLAC and Chile's National Center for Artificial Intelligence, offers a precise snapshot of the AI ecosystem in the region. For…

A critical reading of the Dominican Republic's standing in the Latin American Artificial Intelligence Index (ILIA) 2025 and what it means for our regional competitiveness.
The recent Latin American Artificial Intelligence Index (ILIA) 2025, produced by ECLAC and Chile's National Center for Artificial Intelligence, offers a precise snapshot of the AI ecosystem in the region. For the Dominican Republic, the results present a strategic paradox: absolute progress accompanied by relative decline. This duality deserves a dispassionate analysis that goes beyond the optimistic headlines.
Media coverage has celebrated our classification as an «adopter» country and our score of 44.96 out of 100, above the regional average of 42.98 points. However, a more rigorous reading reveals an uncomfortable reality: The Dominican Republic has fallen from seventh place in the ILIA 2024 to ninth in 2025. This leads us to urge the national press to be more transparent and rigorous when reporting on progress in artificial intelligence.
This two-place drop does not necessarily reflect a deterioration in our absolute capabilities, but something potentially more worrying: other countries in the region are accelerating their investments and building their AI ecosystems with greater speed and strategic coherence. Costa Rica and Peru have overtaken us, which shows that dynamism in this field is not static. Regional competition is intensifying while we keep an insufficient pace of transformation.
Concrete progress in the country
It would be unfair to ignore the tangible progress. The country has shown accelerated improvement in three critical dimensions:
- Digital Connectivity and Democratization of Access: With an internet penetration of 85% in 2024, we have built a solid infrastructure base. The Dominican Republic is among the main users of generative AI in the region, alongside Chile, Peru, Uruguay, Panama and Costa Rica. This technological democratization, enabled by simplified interfaces, allows people without specialized technical training to interact with advanced language models and AI tools.
- Human Talent Development: There is a sustained increase in computer science training programs, although still far from meeting market demand. Universities and technical training centers have begun to incorporate specialized curricula in machine learning, data science and cybersecurity.
- National AI Strategy (ENIA): Since October 2023, the country has had a National Artificial Intelligence Strategy. This institutional framework, though perfectible, sets out guidelines for the responsible adoption of AI and the development of endogenous capabilities.
The CiudadanIA case
The ECLAC report highlights CiudadanIA as an emblematic case of AI application in the Dominican public sector. The project sets up physical interaction points in high-traffic areas, using machine learning models and natural language processing (NLP), including Large Language Models (LLMs), to collect representative data from citizens.
According to the report, the CiudadanIA architecture seeks to train algorithms that offer personalized assistance, information and recommendations, encouraging active citizen participation in the design of a smart government system. This bottom-up approach compares favorably with top-down technology implementations that ignore the real needs of end users.
Although CiudadanIA is a technically competent effort, its impact on the innovation ecosystem deserves analysis through game theory. When the public sector positions itself as the main innovator with virtually unlimited budgets and no pressure for return on investment, it creates a non-cooperative Nash equilibrium where private actors rationally choose not to compete.
The phenomenon is predictable: why would an MSME invest scarce resources in developing AI solutions for smart government when the State is already developing CiudadanIA with financial capacity that no private actor can match? The answer is simple: it wouldn't. The business sector withdraws strategically, hoping the government will solve the innovation problems while they concentrate on activities with lower regulatory risk and asymmetric competition.
This dynamic generates inverse technological dependence: instead of the public sector creating the conditions for private innovation to flourish, it becomes a substitute for it. The result: isolated government projects that do not scale to the productive ecosystem, do not generate sustainable specialized employment outside the public sector, and do not build capacity to export technological solutions.
The uncomfortable question: If CiudadanIA is our flagship case according to ECLAC, why don't we see dozens of Dominican startups competing for government AI contracts? Why is there no vibrant ecosystem of local technology providers feeding on public demand? The answer points to an institutional design that favors direct government execution over market building.
The gap between awareness and investment
A revealing finding from a Microsoft study from early 2020 persists as a structural challenge: 70% of Dominican micro, small and medium-sized enterprises acknowledge that AI will revolutionize the productive sector in the short term, but only 14% invest in it.
The evolution of the AI ecosystem in the Dominican Republic reveals a complex dynamic that requires a conceptual distinction between consumption and investment:
- AI consumption in 2025: According to Tabuga Intelligence, AI consumption in the country is estimated to exceed 2025 US$300 million, embedded as a layer within the software currently distributed to private organizations. This consumption is concentrated especially in:
- Machine Learning for finance: Credit scoring models, fraud detection and risk management
- E-commerce: Recommendation systems, inventory optimization and personalized experiences
- Cybersecurity: Threat detection, anomalous behavior analysis and automated response
- NLP (Natural Language Processing): Conversational chatbots and customer service automation
- Investment in Local AI Development: The investment trajectory in endogenous capabilities presents a different reality:
- 2020: Practically nil investment
- 2024: Approximately $10 million USD
- Projection 2030: $738.77 million USD, according to industry projections cited by Statista.
Although the projection of US$738.77 million for 2030 represents exponential growth compared with the current US$10 million, this figure demands rigorous scrutiny. On what basis is this growth of almost 74 times in six years projected?
There is evidently a “hype” in the global technology ecosystem. Without a historical track record to support such acceleration, and without clearly defined institutional mechanisms to catalyze this investment, we risk confusing aspirations with well-founded projections.
Even more telling: even if the projected $738.77 million is reached, the gap against the $300 million already spent annually on 2025 exposes a structural asymmetry.
We are spending in one year what we project to invest in local development over almost three decades of cumulative effort. This configuration consolidates the Dominican Republic as a consumer market for external AI, not as a generator of its own capabilities.
The GII 2025 warned of it
The decline in the ILIA 2025 is not an isolated phenomenon. The Global Innovation Index (GII) 2025 confirms a deeper deterioration in our overall innovation capacity:
DR's position in the GII:
- 2020: Position 90
- 2022: Position 90 (stagnation)
- 2023: Position 94 (a drop of 4 places)
- 2024-2025: Position 97 (stagnation in mediocrity)
This downward trajectory of seven positions in five years reveals a systematic erosion of our innovation capabilities, attributable not to methodological fluctuations but to growing structural weaknesses.
We Spend, but We Don't Produce
The GII diagnosis 2025 is devastating: «The Dominican Republic produces fewer innovation outputs relative to its level of investment in innovation inputs». In plain language: we invest in education, infrastructure and institutions, but we are not turning those resources into patents, scientific publications, global trademarks or innovative companies.
Critical weaknesses identified:
- Position 138/139 in scientific and technical articles per billion (PPP$ GDP)
- Position 135/139 in trademarks by origin
- Rank 128/139 in industrial designs
- Rank 106/139 in researchers per million inhabitants
- Zero universities in the global QS ranking
- Zero unicorns tech unicorns
- Zero brands Dominican brands among the global top 5,000
Regional comparison
The GII 2025 places the Dominican Republic at rank 12 among 21 economies in Latin America and the Caribbean, behind Puerto Rico, Chile, Mexico, Costa Rica, Brazil, Uruguay, Colombia, Argentina, Panama, Peru and Jamaica.
Particularly worrying: we perform below what is expected for our level of economic development. Countries with a similar GDP per capita produce significantly more innovation. This inefficiency suggests the problem is not primarily one of resources but of institutional allocation, sectoral coordination and incentive design.
The decline in AI (from 7th to 9th place in ILIA) occurs within a broader deterioration of innovation capacities. This suggests that the challenges we face in AI —lack of specialized talent, inadequate infrastructure, a disconnect between academia and industry, no scalable success stories— are symptoms of a systemic disease, not isolated sectoral problems.
The issue is not that «we lack an AI strategy.» The issue is that we lack a functional innovation ecosystem where talent, capital, infrastructure and institutional framework converge to turn knowledge into economic value. Until we attack this root cause, we will keep falling behind on every relevant index.
Systemic challenges
The relative decline in the ILIA 2025 must be understood in the context of deep systemic challenges:
- Inadequate Technological Infrastructure: The lack of robust data centers and national-scale cloud computing platforms limits the development of complex AI projects. Dependence on international cloud solutions raises questions about digital sovereignty and long-term operating costs.
- Brain Drain and Talent Shortage: Specialized AI training remains limited, and emerging talent often migrates to international markets with better pay and professional development opportunities. This drain of human capital restricts the country's capacity to carry out large-scale domestic projects.
- Persistent Digital Divide: Despite an internet penetration of 85%, 40% of the population —concentrated mainly in rural and peri-urban areas— lacks basic digital skills. This digital exclusion creates a social fracture in which the benefits of AI could amplify existing inequalities instead of democratizing opportunities.
- Outdated Regulatory Framework: The absence of specific legislation to combat malicious uses of AI, such as deepfakes, and the lack of rules on data privacy, cybersecurity and ethical use of algorithms create a legal vacuum that inhibits responsible investment and exposes citizens and companies to emerging risks.
- Insufficient multisectoral coordination: Collaboration among universities, the private sector and government remains fragmented. This disjointedness limits the capacity to advance applied research, develop patents and build a cohesive innovation ecosystem. The public sector's eagerness to present itself as innovative to the international ecosystem is obvious, forgetting that “private innovation” precedes government regulation.
Sectoral opportunities
Despite the documented systemic challenges, the Dominican Republic has natural competitive advantages in sectors where AI could catalyze transformations –but only if we urgently implement the recommendations above to build the capabilities we currently lack. Otherwise, these “opportunities” will be captured by foreign companies with imported solutions:
- Tourism: With 31 million Chinese travelers heading to North America each year, AI can position the country to capture a significant segment of this market through advanced service personalization, intelligent recommendation systems and adaptive tourism experiences.
- Logistics and Supply Chain: Back-office automation, predictive operations based on demand analysis, smart logistics assets and new customer experience models can transform the sector's competitiveness.
- Financial Services: Integrating advanced data analytics can deliver more secure payment solutions, improve credit risk management, optimize fraud detection and facilitate financial inclusion through alternative credit scoring models.
- Climate Change Adaptation: As a nation vulnerable to extreme weather events, AI can optimize water management, rationalize post-disaster energy distribution, build intelligent early-warning systems and improve infrastructure resilience through predictive models.
Strategic recommendations
To reverse the relative decline and consolidate a competitive position in the region, a multidimensional approach is required. Here I share a few unsolicited options that I understand to be critical to break the inertia the authorities of the day have sunk us into:
- Aggressive, sustained investment: Increase public and private investment in AI exponentially, setting up tax incentives for companies that adopt AI technologies and foster local R&D.
- Talent training ecosystem: Develop accelerated training programs in AI, machine learning and cybersecurity, in partnership with prestigious international universities. Create incentives to retain specialized talent, including competitive compensation and applied research opportunities.
- A robust and agile sandbox framework:
- The Latin American experience with regulatory sandboxes, extensively documented by the Inter-American Development Bank, shows that these controlled experimentation spaces can speed up innovation without compromising citizen protection. The Dominican Republic successfully implemented this model in the financial sector, allowing fintechs to test products in a supervised environment before final regulation. Why don’t we replicate this success in AI? An Artificial Intelligence Sandbox should:
- Define clear entry criteria: Not only large corporations, but also MSMEs and startups with innovative solutions.
- Set measurable success metrics: Creation of specialized jobs, import substitution in technology, service exports.
- Create risk-proportional protections: Light regulation for low-risk applications (commercial chatbots) and strict oversight for high-risk ones (automated credit systems, surveillance).
- A fixed time frame for experimentation: 12-24 months to validate business models before permanent regulation.
- Mandatory cross-sector coordination: The Superintendencia de Bancos, INDOTEL and the Ministry of Industry and Commerce working together.
- The alternative to the sandbox is the status quo: nonexistent regulation that exposes citizens to risks (deepfakes, algorithmic discrimination) or premature regulation that smothers innovation before it is born. The controlled experimentation model offers the balance between protection and development that our ecosystem urgently needs.
- Mass digital literacy: Implement national digital literacy programs aimed at the 40% of the population without basic digital skills, narrowing the gap that threatens to deepen inequalities.
- National data and computing platform: Develop critical infrastructure of data centers and cloud computing platforms that guarantee digital sovereignty and reduce operating costs over the long term. This goes beyond the innovation strategy, the ENIA and other models that have not attacked the root of the problem.
- National AI Observatory: Establish a Public/Private Artificial Intelligence Observatory that documents success stories, monitors global trends, assesses sector impact and serves as a meeting point for academia, government and industry.
- Success-story strategy: Build on projects like CiudadanIA to create local reference points that are not public projects, to demystify AI adoption by organizations that have “infinite budgets”, showing tangible returns on investment, especially for MSMEs.
A matter of commitment
Setting out an agenda for the next 12 months is a clear commitment, one that allows no excuses from any sector nationwide.
To reverse the inertia in innovation, the business sector urgently needs to form a consortium of 15-20 companies that commit funds to AI innovation under private governance, ending the perpetual wait for government incentives that never arrive.
The authorities should turn CiudadanIA from a state project into a modular public tender executed by Dominican companies, implement a formal regulatory sandbox with published entry criteria and an approval process of at most 30 days, and establish a technology-transfer requirement in every government contract.
Academia must increase mandatory industry certifications in technology curricula, create AI labs that tackle real Dominican problems (hurricane prediction, electricity optimization, fraud detection in remittances), and set up dual university-industry programs in which students work 20 hours a week in the consortium.
Angel investors with patient capital of 7-10 years can join in and increase their openness to emerging Dominican startups.
Next year –as usual, we will publish our analysis comparing these GII and ILIA metrics against actual results, identifying progress or setbacks. Because an innovation ecosystem is built on more than good intentions or speeches; measurable execution, public accountability and clear consequences are the formula to remedy inaction.
The cost of complacency
The fall from seventh to ninth place in the ILIA reflects a deeper crisis: our position 97 in the Global Innovation Index 2025 and a seven-place drop in five years. We are losing ground in AI and in the ability to turn investment into innovation, knowledge into economic value and education into productivity.
AI is already transforming the Dominican Republic. We must decide whether we will be architects of this transformation or consumers of innovation made elsewhere. The decisions we make in the next 18-24 months will determine whether we reverse our downward trajectory or keep depending on external solutions. Complacency will bring greater technological dependence, a transfer of economic value and a growing gap with countries that invest strategically. The time to choose a different path is now; afterwards it will be too late.
Author's note (October 2026): the original version of this article stated that the Dominican Republic ranked sixth in the ILIA 2024 and that Mexico, Peru and Argentina had overtaken it. The correct position was seventh; the drop to 2025 was two places, and the countries that overtook it were Costa Rica and Peru (Mexico and Argentina were already ahead in 2024). The central argument stands.