Will AI take jobs in Jamaica?

Here’s an odd thing about Jamaica in 2026. Unemployment is low, 3.7% in April, even after Hurricane Melissa. And at the same time, the call centre sector, a big employer of young Jamaicans, has been shrinking for two years.

So which is it? Is AI coming for our jobs, or is that just something people in America worry about?

We have a stake in this question. We build AI that answers phone calls and WhatsApp messages for businesses. That’s the same work thousands of Jamaicans do in call centres every day. So we’re going to try to answer it honestly, including the parts that are uncomfortable for us.

The short answer

AI will take tasks first, then some jobs, and the jobs it takes first are the ones built mostly out of repeatable conversations and paperwork.

Most Jamaican work isn’t like that. A mechanic, a nurse, a farmer, a chef, a hotel bartender, a mason: AI can help them, but it can’t do the job. The risk is concentrated in office work, and above all in call centres.

What the big studies say

Every few months a headline says AI will “wipe out” some huge share of jobs. Read the actual reports and the word they use is exposed.

• The IMF estimated in 2024 that almost 40% of jobs worldwide are exposed to AI: about 60% in rich countries, 40% in emerging economies like ours, and 26% in low-income countries. In rich countries, it said, roughly half of those exposed jobs could benefit from AI rather than be replaced.
• The ILO’s 2025 update found about one in four jobs worldwide is exposed to generative AI, with clerical work the most exposed of all. Because women hold more clerical jobs, they face more of the risk.
• For our region specifically, a joint ILO and World Bank study estimated that 26 to 38% of jobs in Latin America and the Caribbean are exposed to generative AI, but only 2 to 5% are at risk of being fully automated. Another 8 to 14% could become more productive.

“Exposed” means AI can do some part of the work. A bank teller whose job is 30% data entry is exposed. That doesn’t mean the teller is gone. It means the 30% is.

Why Jamaica looks different

The IMF made an interesting point about our region in 2025: less than half of jobs in Latin America and the Caribbean are likely to be significantly affected, mostly because so much work here is informal. The higgler, the taxi man, the small farmer, the person fixing phones from a shop front. AI doesn’t have much to replace there.

That sounds like good news, and partly it is. But the IMF also warned that the same thing means we could miss out on the gains. If AI makes a rich country’s office workers twice as productive and ours stay the same, the gap between us grows.

And the jobs that are exposed in Jamaica are some of the best formal jobs we have for young people without a degree: customer service, data entry, back-office processing. Which brings us to the call centres.

The call centre question

Outsourcing became one of Jamaica’s big success stories. Companies from the US, Canada and the UK set up in Montego Bay, Kingston and Portmore because we speak English, share their time zones and cost less.

Now the sector is shrinking. The Observer reported that it has shed about 12,000 jobs in two years, and The Gleaner put employment at about 40,000, down about a third. Interestingly, the head of the Global Services Association of Jamaica told The Gleaner that AI is not one of the causes, pointing instead to low productivity, the damage from Hurricane Melissa, a tight labour market and firms moving to countries with bigger incentives. Industry leaders have also argued that AI will reshape the sector rather than replace it.

We believe them about the past two years. We’re less sure about the next five.

Here’s what we see from the inside. When we build an AI front desk for a business, the calls it handles best are exactly the “tier one” calls that make up a big part of call centre work: What time you open? Where’s my order? How much for delivery? Can I book for Thursday? An AI can answer those at 2 AM, in seconds, for a fraction of the cost of a person.

The IMF said it plainly too: among exposed jobs in our region, call centre roles face a high risk of displacement.

So the honest answer is this. The simple, scripted calls are going. The hard calls are staying for now: the angry customer, the complicated billing problem, the sale that needs persuading, the situation no script covers. A sector built on the simple calls has to move to the hard ones, quickly, or it will keep shrinking.

The jobs most likely to change first

Based on the research and what we see building these systems, these are the jobs to watch:

• Call centre work on scripted, repeat questions.
• Data entry and back-office processing. Typing information from one place into another is what AI does best.
• Receptionists and booking clerks. Answering, scheduling and reminding can all be automated today.
• Basic bookkeeping. Sorting receipts and matching payments is getting automated fast.
• Transcription and simple translation.
• Entry-level writing and design. Product descriptions, simple flyers, social media captions.

Notice what these have in common. None of them disappear overnight. In most cases one person with AI starts doing the work of three, and the other two jobs quietly don’t get replaced when people leave.

The jobs AI can’t touch soon

Work that needs hands, judgement in a messy real place, or trust between people is safe for a long while:

• Trades: electricians, plumbers, mechanics, masons, welders.
• Health care: nurses, care workers, therapists.
• Hospitality: chefs, bartenders, the staff who make a hotel feel like Jamaica.
• Farming and fishing.
• Teaching, coaching and anything built on a relationship.
• Sales that depend on trust.

Robots that can do physical work are coming, but slowly, and they arrive in rich countries’ factories and warehouses long before they arrive on a building site in May Pen.

The jobs it creates

It isn’t only subtraction. The World Economic Forum’s 2025 survey of employers projected about 170 million new jobs and 92 million lost worldwide by 2030, a net gain of 78 million. AI and data specialists are among the fastest-growing roles.

But the new jobs don’t appear in the same place, or for the same people, as the old ones. A 22-year-old in a Montego Bay call centre doesn’t automatically become a machine learning engineer in Toronto. That gap is where policy matters.

What we think should happen

This is an essay, so here’s our opinion.

1. Train for the work around AI, now. Not “learn to code” for everyone. Teach people to use AI tools inside the jobs they already do, and to handle the hard cases AI can’t. The government has said it will roll out an AI workforce-readiness strategy. It should move faster than the job losses, not chase them.

2. Outsourcing firms should move up before they’re pushed. Sell what AI can’t do yet: complex support, collections that need empathy, sales, quality checking the AI itself. A Jamaican agent who supervises ten AI conversations is worth more than one who answers a single call.

3. Small businesses should use AI to grow, not just cut. When a shop’s AI answers every call it used to miss, it gets more customers. More customers can mean more staff, not fewer. That’s the version of this we want to build.

4. Count it honestly. We couldn’t find any official count of how many Jamaican jobs are being changed by AI. The national AI policy being drafted by the National AI Task Force should fix that. You can’t plan for what you don’t measure.

What you can do

If your job is on the “change first” list, don’t panic, and don’t wait.

• Learn the AI tools in your field now, while knowing them is still an advantage.
• Get very good at the parts of your work that need a person: solving the strange problem, calming the angry customer, closing the sale.
• If you’re choosing a career, think about work that combines a skill with people or places, not just a screen.

AI won’t take most Jamaican jobs. It will take the parts of jobs that look the same every day. The people who do best will be the ones who stop doing those parts first.

Sources

1. Jamaica Observer, unemployment at 3.7% in April 2026, June 2026.
2. IMF, AI will transform the global economy. Let’s make sure it benefits humanity, January 2024.
3. UN News, one in four jobs exposed to generative AI, ILO, May 2025.
4. ILO and World Bank, Buffer or Bottleneck? Employment exposure to generative AI and the digital divide in Latin America, 2024.
5. IMF, How artificial intelligence can boost productivity in Latin America, March 2025.
6. Jamaica Observer, the outsourcing sector’s slide, July 2026.
7. The Gleaner, report on outsourcing job losses, with experts on the role of AI, July 2026.
8. Jamaica Observer, AI to reshape, not replace, Jamaica’s global services sector, March 2026.
9. World Economic Forum, Future of Jobs Report 2025, January 2025.
10. The Gleaner, government to implement AI workforce-readiness strategy, March 2025.
11. Office of the Prime Minister, first meeting of the reconstituted National AI Task Force, 2026.