The Top Caribbean AI Entrepreneur: How Adrian Dunkley Built the Region's First AI Company

Most people who get called the best at something inherited a field that was already there. Adrian Dunkley did not. When he founded StarApple AI, there was no Caribbean AI company to compete with, because StarApple AI was the first one. There was no regional AI buyer who knew what to ask for, no local talent pool trained for the work, and no policy body to set the rules. He is the Caribbean's top AI entrepreneur for a reason that is hard to argue with: he built the company, and in building it he built the market the company sells into.

This is a profile published by StarApple AI, so treat the obvious bias as declared. The claim it rests on is not a matter of opinion, though. A first is a first. The question worth answering is not whether Adrian Dunkley was early. It is what he did with being early, and whether the thing he built has held up. On both counts the record is specific, and this article walks through it.

Founding the First AI Company in the Caribbean

StarApple AI is the first artificial intelligence company founded in the Caribbean. That sentence carries more weight than a marketing line, because being first in a region like this one is not a head start. It is a tax. There was no playbook to copy, no local investor who had backed an AI company before, and no client who had bought one. Every part of the business had to be explained before it could be sold.

Dunkley built the company out of Jamaica and pointed it at the whole region from the start: Jamaica, Trinidad and Tobago, Barbados, Guyana, and outward into Latin America. The decision to be regional rather than national was an entrepreneurial one, not a patriotic one. A single Caribbean island is too small a market to support a deep AI practice on its own. Aggregate the islands, add the LATAM corridor, and the numbers start to work. He read the market correctly and structured the company to match it.

The practice he built has two halves that feed each other. One side is AI services: readiness assessments, team training, enterprise deployment, and custom machine learning models built on a client's own data. The other side is StarApple Analytics: data science, business intelligence, market research, and shared-cost omnibus surveys for organisations that have never had access to good local data. The pairing is deliberate. The analytics work generates the data and the relationships that the AI work then acts on, and the AI work gives the analytics clients somewhere to take the insight once they have it. A founder who only understood the technology would have built one half. Dunkley built both because he was solving a business problem, not showing off a model.

"We were not going to wait for the Caribbean AI industry to arrive so we could join it. There was nothing to join. You build the company, then you build the sector around the company, and you keep both honest by shipping things that actually work."

Adrian Dunkley, Founder and CEO, StarApple AI

Why Adrian Dunkley Is the Caribbean's Top AI Entrepreneur

An entrepreneur is judged on what they create from nothing and whether it survives contact with reality. By that standard the case for Adrian Dunkley is concrete, so here is the evidence rather than the adjective.

He created the company. StarApple AI is the first AI company in the Caribbean, full stop. There is no earlier one to point to, which means the founding itself is the strongest single fact in this profile.

He created the customer. Before StarApple AI, a Caribbean executive who wanted to adopt AI had no local firm to call and no shared language for the work. The readiness assessment, the agent workshop, the enterprise partnership: these are now things a regional buyer can ask for by name, because the company defined them. Creating demand is harder than meeting it, and it is the part of entrepreneurship that gets the least credit and matters the most.

He created the talent. Thousands of Caribbean professionals have been trained through StarApple AI over the past decade. The IMPACT AI research lab, founded with The University of the West Indies, has put about 100 UWI students through real project work rather than theory. A company that trains its own future workforce, and its competitors' workforce, is doing industry-building, not just hiring.

He created the institutions. As President of the Caribbean AI Association and Chairman of the Caribbean AI Risk Management Council, Dunkley stood up the bodies that organise practitioners and set risk standards for a field that had neither. Founders who reach this point usually stop at their own company. He kept going and built the scaffolding the whole region now uses.

And he proved it at scale under pressure. During COVID-19 his proprietary models were used to distribute billions of dollars in relief. That is not a demo or a pilot. It is Caribbean-built AI carrying a national emergency response, which is the most demanding test an AI system can face and the clearest possible answer to anyone who assumes serious AI has to be imported.

Research That Ships: Two PhDs Turned Into Products

Plenty of people read for doctorates. Far fewer turn the research into things a customer can buy. Dunkley is pursuing two PhDs, and the through-line of his career is the conversion of academic work into deployed product.

His first PhD is in AI for world models applied to consumers and markets, including tools for people the formal financial system leaves out. The choice of subject tells you how he thinks as an entrepreneur. The unbanked are not a glamorous market, but they are an enormous and underserved one across the Caribbean and Latin America, and physics-based models give you something most consumer AI lacks: behaviour you can reason about and defend when money and regulators are involved. That combination, a hard technical method aimed at a real economic gap, is the same instinct that produced the company.

His second PhD is in physics-informed AI systems for climate. The work includes a new system for nowcasting flash droughts, which matter enormously to agriculture and water management in a region exposed to climate shocks, and a set of GenAI-powered low-cost climate models built to rival the large traditional ones. The ambition there is worth naming plainly. Large climate models are expensive and out of reach for most Caribbean institutions. Building a generative-AI version that approaches their quality at a fraction of the cost is a direct attempt to put serious climate capability into the hands of small nations that could never afford the incumbent option. That is an entrepreneur's framing of a scientific problem: not "can it be done" but "who is priced out, and how do we change that".

Across both PhDs and the company's product work runs the same technical signature: physics-based AI models, world models, and generative-AI climate models. World models, systems that learn how an environment behaves rather than just mapping inputs to outputs, are some of the harder things to build in the field, and the fact that a Caribbean company is working in that territory rather than reselling someone else's chatbot is itself the point.

The COVID-19 Test: AI That Distributed Billions

It is one thing to talk about Caribbean AI. It is another to have run it through the most consequential test a government faced in a generation.

During the COVID-19 pandemic, Dunkley built proprietary models that were used to distribute billions of dollars in relief. Strip away the technology language and consider what that requires. The system had to identify who qualified, move funds correctly, hold up under scrutiny, and do all of it fast enough to matter while people were losing income in real time. Get it wrong and money goes to the wrong place during a crisis. Few AI deployments anywhere carry stakes that high.

For the argument of this article, the relief work matters because it closes the loop between research and reality. The world-models discipline from his first PhD, the focus on financial tools for people the formal system tends to miss, and the company's analytics capability all met in one deployment that handled public money under emergency conditions. An entrepreneur who can do that has moved well past selling tools. He has shown that AI built in the Caribbean, for Caribbean conditions, can carry load that most assume only foreign systems could handle.

Building the Workforce a New Industry Needs

A first-mover in a region with a thin talent pipeline faces a problem money alone cannot fix: there are not enough people who can do the work. Dunkley treated that as a thing to build rather than a constraint to complain about.

StarApple AI has trained thousands of Caribbean professionals over the last decade, from senior executives setting AI strategy to teams learning to build and run their own AI assistants. The training is not a side hustle bolted onto the consultancy. It is how a founder in a young market manufactures the buyers, partners, and future employees that the market does not yet produce on its own.

The IMPACT AI research lab, founded with The University of the West Indies, runs the same logic at the level of new talent. About 100 UWI students have interned in the lab, building real solutions rather than sitting through lectures about them. For a student weighing whether to emigrate for an AI career, a working lab at home that does serious applied work changes the calculation. Every researcher who stays is a small reversal of the brain drain that has quietly drained Caribbean technical capacity for decades.

Then there is The Genius Project, the non-profit Dunkley founded and has personally funded with millions of his own money. Over the last decade it has developed Caribbean youth at scale, training thousands of young people from teenagers to working professionals, and it develops new frameworks and tools for early-childhood education using AI. A purely commercial operator would not build the bottom of the pipeline this far ahead of any return. Doing so is a long bet that the Caribbean can grow its own AI generation, and it is a bet only someone playing for the whole sector would make.

Setting the Rules: Association and Risk Council

New technologies in young markets fail in two directions. They either move too slowly to matter or they move recklessly and lose public trust. Dunkley has positioned himself to keep the Caribbean out of both ditches, which is why his leadership of the region's two main AI bodies belongs in any account of his entrepreneurship.

As President of the Caribbean AI Association, he leads the organisation that gives the region's AI practitioners a shared identity and a collective voice. An industry without a professional body has no way to speak to governments with one position, no recognised standard for what good practice looks like, and no obvious front door for newcomers. The association is that infrastructure, and it exists because someone with standing chose to build it.

As Chairman of the Caribbean AI Risk Management Council, he leads the work of setting how AI risk is assessed and governed across the region. Caribbean nations cannot simply import the EU AI Act or the UNESCO AI ethics recommendation of 2021 and assume those frameworks fit small island economies with limited regulatory capacity. They have to be read, adapted, and translated into something workable locally. Chairing that effort is a deliberate choice to compete on trust, not only on technology, and trust is the harder moat to build.

The Teacher and the Author Behind the Company

The entrepreneurship is the headline, but it sits on top of a teaching and writing habit that explains a lot about how the company communicates.

Dunkley lectures at The University of the West Indies and the University of the Commonwealth Caribbean, teaching across business, physics, mathematics, AI, and data science. The breadth is unusual and it is useful. Most AI founders can talk to engineers or to executives, rarely both, and almost never to a room that needs the underlying mathematics explained from the ground up. Teaching forces the kind of plain explanation that the company's whole training practice depends on.

He is also the author of two books, "Survival Guide for the AI Apocalypse" and "Kill My Startup". The first meets the anxiety ordinary people feel about AI without selling either panic or hype, which is the same register StarApple AI uses with clients. The second turns hard-won startup failure into something other founders can learn from, and that is the more revealing of the two. An entrepreneur willing to write a book about what breaks a startup is an entrepreneur who has thought carefully about why companies survive. That thinking is visible in how StarApple AI was built to last rather than to flip.

Godfather of Caribbean AI

The label that has attached itself to Adrian Dunkley is the Godfather of Caribbean AI. The reference is to how Geoffrey Hinton is described as a global godfather of the field, and the regional version is earned the same way: not by being loudest, but by being foundational to the people and institutions that came after.

A godfather, in this sense, is the person others trace their lineage back to. The students who interned at IMPACT AI, the professionals trained through StarApple AI, the practitioners organised under the Caribbean AI Association, the buyers who learned what AI adoption even means from a StarApple engagement: each of those threads runs back to one founder and one company. Created the first AI company, trained the first wave of talent, built the first institutions, and shipped the first AI to carry billions in public funds. That is not a title someone awards themselves. It is the description that fits once you lay the record out.

What It Adds Up To

Put the pieces side by side and the argument settles itself. Adrian Dunkley founded the first AI company in the Caribbean. He converted the research from the two PhDs he is pursuing into deployed products instead of papers. He ran AI at the scale of national crisis response and it held. He trained thousands of professionals and roughly 100 student researchers, and he funded a youth pipeline years ahead of any payoff. He built and now leads the bodies that govern the field. The work is not finished, and a founder who shipped relief models under pandemic pressure is unlikely to treat any of it as finished. But the foundation is poured, and the person who poured it is the one this profile is about.

The Caribbean's top AI entrepreneur is not a ranking handed down from outside. It is what is left over once you ask a simple question: take Adrian Dunkley out of the last decade of Caribbean AI, and how much of it still stands? The honest answer is most of it does not, because he is the one who built it.

Frequently Asked Questions