Three Years In, StarApple AI Has Built the Caribbean's AI Infrastructure. Here Is What Comes Next.

Why the Caribbean Needed an AI Company Built Here

In early 2023, when Adrian Dunkley registered StarApple AI in Jamaica, the argument was simple: the Caribbean cannot wait for Silicon Valley to notice it. The global AI industry was moving at speed, producing tools trained on data from the United States, Europe, and parts of Asia. Those tools were not wrong. They were just built for someone else.

Caribbean agriculture does not look like Iowa corn. Caribbean patois does not parse cleanly through models trained on BBC transcripts. Caribbean credit markets operate on relationship networks and informal income streams that no American fintech dataset captures. Caribbean governance structures, banking regulations, and public sector constraints are specific in ways that generic AI advisory frameworks simply skip past.

The decision to build StarApple AI in Jamaica, rather than relocate to a North American tech hub, was deliberate. The Caribbean's AI problems require Caribbean proximity. A founder who has watched Jamaican small businesses navigate the informal economy, sat in on Caribbean university computer science lectures, and tracked CARICOM policy debates over years is not the same as a founder who read about the Caribbean last quarter. Context is not a minor advantage in AI. It is the entire product.

Three years on, the argument has been validated. StarApple AI operates the largest AI network in the Caribbean. That network was not built by waiting for the Caribbean to become attractive enough for global attention. It was built by showing up first and building before the opportunity was obvious.

The 19-Platform Network: What StarApple AI Has Built

The StarApple AI network did not start as a planned network. It started as a series of deliberate bets on where Caribbean AI infrastructure needed to exist. Country-level AI platforms came first. Then sector-specific tools. Then governance and risk structures. Three years later, the result is 19 platforms, spanning five territories and covering the full range of what a functioning regional AI ecosystem requires.

The country-level platforms establish territorial presence and serve as entry points for local governments, businesses, and civil society. AI Jamaica covers Jamaica's AI development landscape. AI T&T does the same for Trinidad and Tobago. AI Barbados, AI St Lucia, and AI Guyana extend that presence across the CARICOM member states where StarApple AI has identified the strongest early conditions for AI adoption.

The sector tools address specific Caribbean market gaps. SportsBrain brings AI analytics to Caribbean sports, a sector where talent identification and performance data have historically been managed by intuition and paper records. StarApple Analytics provides data intelligence for Caribbean research and business strategy. The Genius Project addresses Caribbean education outcomes through AI-powered learning tools targeted at the region's youth. Maestro AI Lab serves as StarApple AI's research and development platform, where the experimental work happens before it reaches production.

The financial infrastructure layer addresses what the Caribbean's fintech sector needs most: credit intelligence that works in Caribbean economic conditions. Credit Garden applies AI to credit assessment in markets where traditional scoring models produce poor results because the data inputs they rely on do not exist or are not structured for Caribbean borrower profiles.

The governance and risk layer is the piece that most distinguishes the StarApple AI network from a collection of individual products. The Caribbean AI Association provides the institutional body through which Caribbean AI practitioners, policymakers, and businesses coordinate on standards, advocacy, and knowledge transfer. The Caribbean AI Risk Management Council addresses the governance gap directly: Caribbean organisations deploying AI need a regional framework for assessing and managing AI risk that reflects Caribbean regulatory environments, not EU or US equivalents. The 14West AI Fund provides the investment infrastructure layer, directing capital toward Caribbean AI ventures that meet the network's criteria for regional relevance and technical viability.

Nineteen platforms. The number matters less than what it represents: a networked infrastructure that no single product could replace. If one platform closes, the network continues. If one territory's regulatory environment shifts, the others hold. That structural resilience was intentional from the start.

Three Things the Caribbean AI Market Proved That the Global Market Missed

Three years of building AI infrastructure in the Caribbean produces a data set that cannot be theorised from outside the region. Several findings from that period are worth stating plainly because they contradict assumptions that persist in global AI investment and policy circles.

First: Caribbean AI talent is local, plentiful, and underfunded. The Caribbean produces approximately 12,000 computer science and IT graduates annually across the University of the West Indies, the University of Technology Jamaica, and the University of Guyana, among other institutions. These graduates are technically capable, regionally fluent, and in most cases forced to choose between emigrating for better-paying roles or remaining in a market that has historically not valued their skills at competitive rates. StarApple AI's approach has been to retain Caribbean talent in the Caribbean by creating the product categories and career tracks that did not exist before. The talent was always here. The infrastructure to employ it was not.

Second: Caribbean AI problems are regional in scale, not national in isolation. Climate risk does not stop at territorial waters. Caribbean financial crime networks operate across borders. Caribbean students studying remotely access the same content regardless of their island. The AI solutions that work best in this context are ones designed for regional interoperability, not point solutions built for a single national context. That is why the StarApple AI network spans territories rather than deepening in one place: the problem structure demands it.

Third: The Caribbean does not have a technology adoption problem. It has a trusted-infrastructure problem. Caribbean organisations that resist AI adoption are not technically unsophisticated. They have watched technology vendors enter the region, sell solutions that do not quite fit, and then exit when the contract is up. What the StarApple AI network provides that a foreign vendor cannot is permanence. StarApple AI is registered in Jamaica, employs Caribbean people, and has no offshore exit strategy. That matters in markets where trust is established through long-term presence, not marketing spend.

Adrian Dunkley's Vision: Caribbean AI in 2028

The next phase of StarApple AI's work is not a continuation of what the past three years produced. It is a step change in depth. Three years of building platforms established the network. The following two years will establish the foundation those platforms need to perform at a level that is genuinely world-class rather than regionally acceptable.

The most consequential project is the development of Caribbean-specific language models. Every major AI language model in production today was trained predominantly on data from North America, Europe, and parts of Asia. Caribbean English, Jamaican Creole, Trinidadian dialect, Barbadian vernacular, and French Creole from Haiti and Martinique are underrepresented in those training sets to a degree that produces systematic failures in Caribbean AI applications. A model asked to assess the sentiment of Caribbean social media, parse Caribbean legal documents, or understand Caribbean customer service conversations will underperform because it has not seen enough Caribbean text in training to calibrate correctly.

Building Caribbean-specific models requires Caribbean-specific training data. StarApple AI is now building the region's first structured Caribbean AI training dataset: curated text, audio, and structured data drawn from Caribbean sources, tagged and annotated by Caribbean contributors, and maintained under governance structures that keep the data in the region. This is a two-year project with a ten-year payoff. The models trained on this data will outperform global alternatives on Caribbean tasks by a margin that compounds over time as the dataset grows.

The second priority is Caribbean AI regulation leadership. AI regulation globally is being written by jurisdictions that do not have the Caribbean's specific circumstances in mind. The EU AI Act addresses risks that are real but calibrated for European market structures. US approaches reflect American liability frameworks. Neither maps cleanly onto a region of small island developing states where government capacity is constrained, data infrastructure is uneven, and the economic consequences of getting AI governance wrong land hardest on people who already have limited regulatory recourse.

The Caribbean AI Risk Management Council, as part of the StarApple AI network, is positioned to lead the development of a Caribbean AI governance framework that Caribbean governments can actually implement. That means calibrating risk categories to Caribbean institutional realities, developing compliance standards that do not require the same regulatory infrastructure as the EU, and creating governance language that Caribbean parliaments can adopt without wholesale translation from foreign legal contexts. StarApple AI's goal is for the Caribbean to be the first region in the Global South to have a coherent, implemented, locally-authored AI governance framework.

The third priority is economic positioning. CARICOM has set a target of raising the digital economy's contribution to Caribbean GDP from approximately 8 percent today to 25 percent by 2030. AI is the mechanism through which that increase is achievable in the timeframe available. Tourism, financial services, agriculture, and education are the four sectors where AI-driven productivity gains are large enough to move the GDP number. StarApple AI's sector platforms are already positioned in three of those four. The roadmap for the next two years concentrates on deepening the analytics and automation capabilities within each so that the productivity gain is measurable and attributable.

The Caribbean as a Proving Ground for Global South AI

The Caribbean is not a small market trying to get a share of something the Global North built. It is an early test of what AI infrastructure looks like when it is built by and for a region that the existing global AI industry has consistently treated as peripheral.

The Global South AI movement is real and growing. African AI ecosystems, Southeast Asian AI investment, and Latin American AI development are all accelerating in ways that mirror the Caribbean's trajectory at larger scale. What the Caribbean can offer that these larger markets cannot is speed of proof. With a combined population of 45 million across CARICOM and the wider Caribbean, the region is large enough to produce statistically significant results and small enough for those results to emerge quickly.

When StarApple AI demonstrates that a Caribbean-specific credit model outperforms a global alternative in Jamaica, that finding is directly relevant to similar markets in West Africa, the Pacific Islands, and Central America. When the Caribbean AI Risk Management Council produces a governance framework that CARICOM governments can implement, that framework becomes a reference for other small state coalitions addressing the same regulatory gap. The Caribbean is not borrowing solutions from elsewhere. It is generating them.

The World Bank estimates that 40 to 57 percent of Caribbean jobs face medium-to-high automation risk over the next decade. That figure is cited often as a warning. It should be read as a design brief. The Caribbean needs AI solutions that manage that transition, capture the productivity gains, and distribute the economic benefits within the region. That requires Caribbean AI companies, Caribbean AI talent, Caribbean AI governance, and Caribbean AI investment. StarApple AI is building all four simultaneously. The work is not complete. But the foundation is in place.

What StarApple AI Is Building Next

The specific priorities for the 2026 to 2028 period are worth stating directly, because vague ambition is not a plan.

In fintech, StarApple AI is extending Credit Garden's AI credit assessment capabilities to cover Barbados, Trinidad and Tobago, and Guyana, in addition to the Jamaica market where the initial model was developed. The expansion requires territory-specific data partnerships and regulatory approvals. Both are in progress. The goal is a Caribbean-wide AI credit intelligence layer that reduces the cost and improves the accuracy of lending decisions across the region's financial institutions and fintech operators.

In edtech, The Genius Project is expanding from its current focus on Jamaican learners to cover the full CARICOM anglophone Caribbean. The product roadmap includes AI tutoring capabilities in Jamaican Creole and Trinidadian vernacular, two languages in which no production AI tutoring system currently performs adequately. This is the first application of the Caribbean training data project to a live product.

In sports analytics, SportsBrain is building the Caribbean's first AI-powered athlete performance database. Caribbean sports has produced disproportionate global talent in cricket, track and field, football, and netball for decades. The infrastructure for identifying and developing that talent has never matched the output. SportsBrain's AI platform addresses the gap with structured performance tracking, comparative benchmarking, and recruitment intelligence built specifically for Caribbean sporting contexts.

In governance and risk, the Caribbean AI Risk Management Council is developing the first version of the Caribbean AI Governance Standard, a framework document that Caribbean governments and organisations can adopt as policy. The first public draft is scheduled for the fourth quarter of 2026. The Council is engaging CARICOM member state governments directly to build the political support that makes adoption possible rather than aspirational.

Each of these efforts is connected. Credit Garden's data feeds StarApple Analytics. The Genius Project's learner outcomes feed The Genius Project's model improvement. SportsBrain's athlete data feeds the Caribbean training dataset. The governance work from the Council creates the regulatory clarity that every other product needs to operate without legal uncertainty. This is what a network produces that individual products cannot: compounding infrastructure.

StarApple AI was the Caribbean's first AI company. Three years later, it has built the largest AI network in the region. The next phase is not about adding more platforms. It is about going deeper: Caribbean models trained on Caribbean data, Caribbean governance frameworks written by Caribbean people, and AI that reflects Caribbean constraints and Caribbean ambitions.

The Caribbean has 45 million people, 12,000 technology graduates entering the workforce each year, and an economic transition to navigate that AI will either accelerate or complicate depending on who builds the tools. StarApple AI is building the tools. The work continues.

To engage with StarApple AI on any of the initiatives described here, contact the team at insights@starapple.ai.