AI Literacy in the Caribbean: What Every Citizen Needs to Know by 2027

The most important question facing the Caribbean in the AI era is not which model we use. It is whether the average Caribbean citizen can hold a sensible conversation about what AI is, what it can do, what it should not be allowed to do, and how to make it useful in their own life. That capability has a name. We call it AI literacy, and across the region it is the difference between a population that shapes its technology and one that is shaped by it.

This piece sets out what CAIA means by AI literacy, why we believe it is the foundational public investment of this decade, and the practical curriculum we are putting into the hands of Caribbean ministries, schools, employers, and community organizations. It draws on the work of our country chapters in Jamaica, Trinidad and Tobago, Barbados, Guyana, the Bahamas, and across the OECS, and on the lived experience of the parents, teachers, and workers who have helped us refine it.

What AI Literacy Is, and What It Is Not

AI literacy is often confused with two adjacent things it is not. It is not coding. A Caribbean citizen does not need to write Python to be AI literate, any more than a citizen needs to wire a circuit board to be literate in electricity. And it is not the consumption of think pieces about superintelligence. Those debates have their place, but they are not what an accounts clerk in Bridgetown needs to do her job better next Tuesday.

AI literacy is the practical capacity to do four things. First, to recognize when AI is involved in a decision, a piece of content, or a service you use. Second, to judge whether the AI output in front of you can be trusted, and to know what to do when it cannot. Third, to operate the common tools well enough to add value in your work and your home life. Fourth, to participate meaningfully in the public conversation about how AI should be governed in your country.

That is the bar. It is achievable for every Caribbean citizen from Standard 4 upward with the right materials, the right teachers, and a region that takes the task seriously.

The Four Layers of Caribbean AI Literacy

The Caribbean AI Literacy Programme that CAIA publishes is organized around four progressive layers. They build on each other, but each layer is independently useful.

Layer one, awareness. The citizen can identify AI in the products and services they already use. They know that the autocorrect on their phone, the recommendation feed on their social app, the chatbot on the bank website, and the voice on the customer service line are all AI systems. They know there is no human reading every WhatsApp message at the other end. This sounds basic. It is not. Many adults across the region still hold a mental model of computing that pre-dates 2022, and that gap drives both misplaced trust and misplaced fear.

Layer two, evaluation. The citizen can look at an AI output and ask the right questions. Is this true? How would I check? What incentive does the system have to mislead me? What did it not tell me? This is the same critical reasoning Caribbean students have always been taught to apply to a newspaper article or a politician's speech. Applied to AI, it is the most important defensive skill of the decade.

Layer three, use. The citizen can prompt the common tools to get useful output. They know how to ask a model for a structured answer, how to give it the right context, how to push back when it is wrong, and which tool fits which job. They know that NotebookLM is for studying from sources, that ChatGPT Learning Mode is for being tutored, that Gemini Canvas is for working alongside the model on a document, and that a quick search on Perplexity often beats a long chat with any of them.

Layer four, participation. The citizen can engage with the public debate about AI in their country. They can read a policy consultation document on facial recognition in Bridgetown or generative AI in Kingston schools and form an opinion that is grounded in something more than a headline. Democracy in the AI era depends on this layer existing in numbers.

Where the Caribbean Is Right Now

Honest assessment is the starting point. The CAIA Caribbean AI Literacy Index, which we publish annually, suggests that in 2026 roughly one in three working age Caribbean citizens has reached layer two, roughly one in five has reached layer three, and fewer than one in ten has reached layer four. Variation across countries is wide. Barbados, Trinidad and Tobago, and the Cayman Islands lead on layer three, helped by high literacy rates and active employer training. Jamaica leads on layer four because of its lively press and an unusually engaged civil society. Haiti and the smaller OECS islands trail across the board, reflecting connectivity gaps more than capacity.

The headline is that the region is mid-curve, not flat-footed. We have enough early adopters, enough trained teachers, and enough public interest to move the population as a whole if we organize the effort.

What to Teach, Where

The Programme is built so that each setting can pick the parts that fit. In primary schools, layer one fits naturally into existing computer studies and social studies units. Teachers in Spanish Town, Kingston, and Saint George's have used our Grade 5 module to teach a class to spot AI-generated images, AI-written news, and AI in their own phones. The module takes three lessons.

In secondary schools, layer two and the beginning of layer three sit well alongside CSEC English and Information Technology. The CSEC alignment is deliberate. A child who graduates secondary school in 2027 should leave with the ability to write a prompt, evaluate an output, and understand the basics of how a language model works. CAPE Information Technology will need a deeper revision, and CAIA is in conversation with CXC on what that looks like.

In tertiary settings, layer three is the focus. The University of the West Indies has begun integrating practical AI use across faculties beyond computer science, and CAIA is supporting that work with a faculty development cohort. Caribbean tertiary students who graduate without practical fluency in the tools their employers will use are entering the workforce with a disadvantage that compounds each year.

In workplaces, the entry point is task based. A call center in Montego Bay does not need a six week course on transformer architecture. It needs a two day workshop that teaches every agent to use the AI assistant in their queue, to escalate when the model is uncertain, and to recognize when a customer is trying to manipulate the system. CAIA partners with employers across the region to deliver these workshops at scale.

In communities, libraries and faith institutions are the surprise winners. The most successful CAIA literacy session of 2025 was a Saturday morning workshop at a public library in Castries that drew 80 grandparents who wanted to know what their grandchildren were doing with ChatGPT. The session ran twice and the head librarian asked for a permanent monthly slot. That is the kind of demand that should shape public investment.

What Governments Should Fund

The most leveraged public investment a Caribbean government can make in AI right now is not a data center or a national model. It is teacher training. A trained primary school teacher reaches 30 children a year for the rest of their career. A trained secondary school teacher reaches 150. Train a thousand Caribbean teachers in the next two years and the literacy curve shifts for a generation.

The second most leveraged investment is connectivity at the household level for low income families. The literacy programme assumes a device and a connection. Where those are absent, no curriculum closes the gap. The OECS Commission's work on subsidized connectivity is the kind of regional initiative that should be expanded.

The third is procurement standards. Any AI system bought by a Caribbean government, including in education, health, justice, and welfare, should meet a minimum transparency and bias testing standard. CAIA has published a model procurement clause that small Caribbean states can adopt without retaining a large policy team.

The Cultural Dimension

AI literacy in the Caribbean is not a copy paste of literacy programmes designed in Helsinki, Singapore, or Silicon Valley. Our languages, our humor, our relationship to authority, our oral traditions, and our histories all shape how AI lands here. A model that handles standard English well but misreads Patois as broken English will quietly fail every Jamaican household it touches. A literacy programme that assumes a quiet study desk in a single family home misses the reality of the Caribbean kitchen table.

The Caribbean AI Literacy Programme is designed for our context. It uses Caribbean examples, Caribbean voices, and Caribbean languages where the tools support them. It teaches the limits of those tools where they do not. And it is built to be adapted by each country chapter to fit its particular curriculum, languages, and culture.

What Comes Next

By 2027, CAIA's goal is that the Caribbean reaches a layer three majority. That is, more than half of the working age population in CAIA member countries is competent in using the common AI tools to do part of their work or their household life. We believe this is achievable. The Caribbean reached near universal basic literacy within two generations of independence, against far harder odds than the ones we face today.

This is the work CAIA was founded to do. It is the work our members do in classrooms, libraries, offices, and ministries every week. It is the work that will determine whether the Caribbean enters the second half of this decade as a participant in the AI economy or a passenger in it. We choose participant. We invite every Caribbean person who agrees to join us.