This post is an expanded version of the original essay published on Medium in February 2026. Read the original on Medium ↗
The word "apocalypse" is doing a lot of work in conversations about AI right now. Every week there is a new piece forecasting the end of knowledge work, the collapse of the middle class, the dawn of something either utopian or dystopian depending on which newsletter you read last. The discourse has become a kind of ambient terror - not specific enough to act on, not calm enough to ignore.
I want to do something different. I want to be specific. Because if the AI disruption is coming - and I believe a version of it is, though not the cinematic one - then the useful question is not "will it happen?" It is "who does it hit first, where does it hit hardest, and what does intelligent preparation actually look like?"
I have spent 15 years building AI. I founded the Caribbean's First AI Company, StarApple AI. I have trained over 500 professionals in AI and data analytics. I have sat in rooms with governments across the Caribbean trying to answer exactly this question on a national scale. And I want to give you the honest version of what I have found.
The AI Disruption Is Not What You Think
The story most people are told is about robots taking jobs. Physical automation replacing physical labour. This happened in manufacturing. It is happening in warehousing. It will happen in more sectors. But this is not the primary disruption of the current AI wave. The current disruption is cognitive.
Large language models can write. They can code. They can analyse. They can summarise, strategise, explain, translate, generate images and audio and video. They can do many of the tasks that defined "knowledge work" for the last century of economic development. Not all tasks. Not all of them well. But enough of them, well enough, that the economics of many knowledge work roles are being profoundly altered right now.
The question is not whether this changes things. It does. The question is the pace and the distribution of that change - who it hits first, who it hits hardest, and who it creates opportunity for.
Who Gets Hit First
Based on everything I have seen and studied, the early disruption is concentrated in:
High-volume, routine cognitive tasks. Data entry, report generation, basic coding, template-driven writing, transcription, customer service scripts, first-level legal document review, rote financial analysis. These are tasks where AI is already performing at or above the level of an entry-level or mid-level human worker, at a fraction of the cost and with no need for benefits, management, or time off.
The entry-level gateway. This is the disruption that worries me most for the Caribbean. Historically, entry-level positions in law, finance, consulting, journalism, and software development served as the apprenticeship layer - the place where young professionals learned how to do the work by doing a lot of the work. AI is compressing that layer. A senior lawyer with Claude can do the document review that previously required four junior associates. The junior associates do not disappear. The jobs do.
Geography and access. The disruption is not evenly distributed. Countries with strong AI literacy, digital infrastructure, and adaptive institutions will navigate it better. Countries without those foundations will be hit harder and have fewer tools to respond. The Caribbean is in a complicated position here - talented people, genuine creativity, but uneven infrastructure and a lagging policy response.
Who Survives - and Who Thrives
Let me be clear about something: I do not think the AI disruption ends in mass unemployment. I think it ends in massive restructuring - which is painful and can produce mass suffering during the transition, but is not the same thing as a permanent collapse of human economic value.
The people who thrive in the AI era share certain characteristics that I have observed consistently across the 500+ professionals I have trained and the dozens of organisations I have advised.
They have genuine domain expertise. Not just knowledge - deep, contextual, experience-based expertise that allows them to evaluate AI output, to know when the system is wrong, to ask the right questions of the right tool at the right moment. AI amplifies competence. It does not manufacture it. The expert with AI becomes exponentially more productive. The person without expertise becomes more efficiently wrong.
They are comfortable with uncertainty. The AI landscape is changing faster than any individual can fully track. The professionals who survive are not the ones who have figured out exactly how AI will change their field. They are the ones who have built the cognitive flexibility to keep updating their model as the technology evolves.
They are relentless learners. This is not a new observation, but the AI era makes it more urgent. The half-life of specific technical skills is shrinking. The premium on learning how to learn - on the ability to pick up new tools, new frameworks, new workflows - has never been higher.
They have irreplaceable human skills. Empathy. Political navigation. Creative vision. The ability to build trust with another human being in a room over time. Leadership under uncertainty. These are not skills that AI will automate any time soon. The professionals who have invested in developing them are building a moat that the next model release cannot cross.
The Caribbean's Specific Challenge
I have to be honest about the Caribbean dimension of this, because it matters enormously and because I have been watching it play out for years.
The sectors that employ the largest numbers of educated Caribbeans - BPO and business services, financial services, government administration, tourism, education - are precisely the sectors where AI is creating the most significant medium-term disruption. The BPO sector, which employs tens of thousands across Jamaica, Trinidad, Barbados, and the broader region, is directly in the path of AI automation. Customer service AI, back-office automation, and AI-powered data processing are already displacing BPO roles globally. The Caribbean cannot assume it is immune.
What the Caribbean can do - what it must do - is get ahead of this transition rather than react to it. That means:
- Aggressive AI literacy investment - not optional, not peripheral, but central to workforce development strategy at the government level.
- Sector diversification - building AI-native industries that are creators of AI value rather than only consumers of AI tools.
- Education system reform - preparing students for a world where the ability to collaborate effectively with AI is as fundamental as literacy was in the 20th century.
- Social protection planning - building the safety nets for the workers who will be displaced before the new roles materialise, because the transition will not be painless.
What Efficient Preparation Actually Looks Like
The title of this piece promises efficiency. Let me deliver on it with specifics.
At the individual level: start now. Not "start thinking about it." Start actually using AI tools in your work. The learning curve for working effectively with AI is real and it takes time. Every month you wait is a month of compounding disadvantage relative to the people who are already building the habit. Five hours of deliberate AI use per week for six months will change how you work more profoundly than almost anything else you can do with that time.
At the organisational level: the question is not "will AI change our business?" It is "where in our business will AI create value first, and how do we capture that value before our competitors do?" Every Caribbean organisation, in every sector, should have an AI strategy. Not a technology strategy. An AI strategy - one that maps specific business processes to specific AI capabilities and builds a plan to deploy them responsibly over the next 18 months.
At the government level: the window for getting ahead of this is closing. The AI disruption is not a 20-year horizon problem. It is a 5-year horizon problem for many sectors. Policy responses that were appropriate in 2022 are no longer adequate. National AI strategies need to be enacted, not just drafted. Workforce transition programmes need to be funded, not just proposed. Caribbean governments that move now will be able to shape this transition. Those that wait will only be able to react to it.
A Note on the "Apocalypse" Framing
I use the word "apocalypse" deliberately, and I want to close by explaining why. The original Greek meaning of the word is not "catastrophe." It is "revelation" - an unveiling, a making visible of what was previously hidden.
The AI disruption is an apocalypse in that original sense. It is revealing the fragility of economic structures that depended on the high cost of cognitive labour. It is revealing which skills are genuinely rare and valuable and which ones were valuable only because they were time-consuming for humans to perform. It is revealing which institutions are adaptive and which are brittle. It is revealing which societies have built the resilience to navigate transformation and which have not.
This revelation is uncomfortable. It is also, if you are paying attention, an extraordinary opportunity. The Caribbean can use this moment to build an AI ecosystem that serves our people - that creates genuine value, genuine employment, and genuine economic self-determination - rather than importing AI from elsewhere and hoping it serves us adequately.
That is the work. Not efficient panic. Efficient preparation. And the preparation starts now.
This is an expanded version of the essay originally published on Medium. Read and share the original here ↗