Will AI Take My Job? A Straight Answer for Jamaican Workers

Probably not this year, and probably not the way you are picturing it. Anthropic's labour market research found no systematic rise in unemployment among highly exposed workers since late 2022. Three years into the generative AI era, the mass displacement event has not arrived on the schedule the headlines promised.

But that is not the same as nothing happening. The pressure is real, it is just landing somewhere more specific than most people expect.

What the evidence actually shows

Three findings matter, and they pull in different directions.

Employment has held up. No measurable unemployment surge among exposed workers. If you are established in a role, AI is currently a tool rather than a threat.

Growth is slowing where exposure is high. The same research found roughly a 0.6 percentage point decline in projected employment growth per 10 percentage points of increased AI coverage. Jobs are not vanishing. They are being created more slowly. That is a quieter problem and a more durable one.

The young are absorbing the hit. There is evidence of slower hiring for workers aged 22 to 25 in highly exposed occupations. So the problem right now is not keeping a job. It is getting one, or entering a new field. Fresh out of college trying to get that first job is the real issue.

Globally, the World Economic Forum projects 170 million new roles and 92 million displaced by 2030, with 39% of core skills changing. The net figure is positive. The churn figure is enormous. It is the churn that lands on an individual career.

Jamaica lost 20,000 jobs, and AI did not cause it

This part matters, because a lot of local anxiety is attached to a story the data does not support.

Jamaica's BPO sector fell from roughly 60,000 to 40,000 jobs over two years. Real losses, real households. But when the Jamaica Gleaner put the AI question to industry leadership in July 2026, Yoni Epstein, president of the Global Services Association of Jamaica and CEO of itel, was blunt: "The job loss has been caused by a few things, but artificial intelligence isn't one of them." He named productivity, Hurricane Melissa, labour availability and competing incentives in other jurisdictions.

STATIN's April 2026 figures show unemployment at 3.7% and youth unemployment at 11.7%, with the biggest losses in agriculture, real estate and retail. That reads like a hurricane and a soft economy, not an AI displacement signature.

So why does it feel like AI is taking Jamaican jobs? Because two things are happening at once. Jobs are being lost for ordinary economic reasons, and separately, AI is changing what the surviving jobs require. Blaming AI for the first means preparing badly for the second.

Which jobs are exposed, and which are barely touched

Exposure is measured by how much of a role's real task content AI already handles in professional use, not by how futuristic the job sounds.

That last row is the finding most people miss. Roughly 30% of workers show effectively zero AI coverage, not because the work is simple but because it is physical, situational and happens in places nobody ever documented.

An electrician tracing a fault in an older Kingston building is combining incomplete information, physical inspection, safety judgement and improvisation. No current model does that. Meanwhile the AI boom itself is driving demand for exactly those skills, because data centres need people who can pull cable and keep cooling systems running. That is the argument behind my book, The AI-Proof Career: Why Skilled Trades Are The Future of Work, and the evidence has moved in its favour since publication.

The four things AI cannot replace

Not job titles. Properties.

1. Physical presence in environments that were never designed to be machine-readable
2. Accountability a named person has to carry, sign for, or be licensed to hold
3. Relationship and trust built over time, which in Jamaica closes in a WhatsApp conversation with someone specific
4. Judgement under genuine ambiguity, which is different from analysis

The inverse tells you where risk sits: screen-based, well-documented, repeatable work judged on output volume rather than on who produced it.

How to measure your own exposure

Stop asking whether AI will take your job. That is the wrong unit. Jobs are bundles of tasks, and AI takes tasks. A job only disappears when enough tasks go that the remainder is not worth a salary.

There are two ways to get your number. Both work. One takes twenty minutes, the other takes five.

Option one: do it by hand

Take a sheet of paper and list every recurring task in a normal week, with rough hours. Mark each against the four properties above. High exposure if it is screen-based, documented and repeatable. Low if it needs presence, accountability, relationship or judgement.

Add up the high-exposure hours as a percentage of your week. That number is your actual risk profile, and it is more accurate than any article, this one included.

If it is 60% or more, you have something to act on within two to three years. It will most likely arrive as a role that quietly does not get backfilled rather than a redundancy letter.

If it is around 20%, you have leverage. Automate those hours and move them into the work only you can do.

Most people land somewhere in between, partly protected and partly exposed. That is the normal result and it is far more useful than a yes or no.

Option two: let the diagnostic do it

Why Jamaicans are better placed than they assume

Only 6% of Jamaicans have formal AI training, per the SALISES study at UWI Mona. Locally you are early, not late. Two to six weeks of applied practice tied to your real work moves you from unfamiliar to useful.

The pattern in every labour transition on record is the same. The tool does not replace the worker. The worker using the tool replaces the worker who is not.

Frequently asked questions

Will AI take my job in Jamaica?

For most workers, not in the near term. No systematic unemployment rise has appeared among highly exposed workers since late 2022. The clearest genuine risk is slower entry-level hiring in exposed fields.

Did AI cause Jamaica's BPO job losses?

No. Industry leadership attributed the loss of roughly 20,000 jobs to productivity, Hurricane Melissa, labour supply and incentive competition.

Which jobs are safest from AI?

Skilled trades, healthcare, care work and licensed professions. About 30% of workers show effectively zero AI task coverage, concentrated in physical and hands-on roles.

Am I too old to start learning AI?

No, and experience helps. Knowing when AI output is wrong requires domain knowledge a younger colleague does not have yet. With only 6% of Jamaicans trained, the local bar is low.

How do I know if my specific job is at risk?

Audit tasks, not the job title. List your recurring weekly tasks with hours and mark each for AI exposure. The percentage of your week in high-exposure tasks is your real risk profile. The free diagnostic at www.aiproofcareershub.com does the same thing in five minutes.

Should young Jamaicans still study technology?

Yes, with a caveat. The entry-level squeeze is the clearest displacement signal in the data. Pair technical study with something AI cannot do: a licence, a physical skill, a specialisation or a network. HEART/NSTA trade certification is a serious option, not a fallback.