Artificial intelligence is often discussed in Barbados in terms of technology, productivity and innovation (see here and here) As a Barbadian, and a technologist, there’s a question that I think may need further in-depth analysis:
What happens when AI meets the structure of the Barbadian labour market?
Note: This article is intended to explore and encourage discussion about the potential impact of AI on the Barbados labour market. It is not intended to be a formal study or comprehensive analysis (yes, AI was used to assist with the creation of this article).
According to the Barbados Statistical Service’s 2025 Labour Force Survey, approximately 126,300 people were employed in Barbados. Around 14,000 were clerical workers, 14,900 were professionals, 13,000 were technicians and associate professionals, and 12,200 were legislators and senior officers.
Together, those four broad occupational categories account for approximately 43% of employment. That does not mean 43% of Barbadian workers are about to lose their jobs to artificial intelligence. However, it shows how much of the country’s employment involves occupations in which information processing, communication, administration, analysis and other cognitive tasks play important roles. And those are precisely the areas where generative AI is advancing rapidly.
All Jobs Do Not Face The Same Risk
Consider two workers: One spends much of the day preparing reports, answering routine correspondence, processing documents, entering information into systems and summarising information. The other is a carpenter working on a construction site (related note: the island is currently going through a construction boom as well). Today’s generative AI can already perform portions of the first worker’s activities, but it cannot pick up and use a circular saw or build a roof (for now).
This creates an unusual feature of the AI revolution. Previous waves of automation were frequently associated with machines replacing physical labour. Generative AI has turned some of that assumption upside down. Many white-collar occupations (i.e., office workers) may be more exposed to the current generation of automation than traditional trades.
Barbados provides an interesting case study. In 2025, 12,600 Barbadians were employed in craft and related occupations, while another 7,400 worked as plant and machine operators and assemblers. Those jobs will likely be affected by technology. AI can assist with design, estimating, scheduling, inventory, procurement and numerous other activities surrounding physical industries. But much of the physical work remains considerably harder for generative AI to automate for now.
The Situation Is Different For Many Office Occupations.
Barbados has 14,000 clerical workers. Research from the International Labour Organization (ILO) consistently identifies clerical occupations as having the greatest exposure to generative AI. Its original global study estimated that 24% of clerical tasks had high exposure to automation and another 58% had medium-level exposure. The ILO’s updated 2025 occupational exposure index reached a similar conclusion: clerical occupations remain the most exposed occupational group, although exposure has also increased among some highly digitised professional and technical occupations.
The Barbados numbers reveal another dimension to the issue, one involving gender. Of the country’s approximately 14,000 clerical workers, 11,800 were women. In other words, roughly 84% of Barbados’s clerical workforce was female in 2025. That is particularly noteworthy because international ILO research has also found that women tend to have greater exposure to generative AI because of their greater representation in clerical and administrative occupations.
The potential effects of AI disruption may not be evenly distributed across the Barbadian workforce. There is an important distinction between a job being exposed to AI and a job being replaced by AI. An accountant, lawyer, engineer, financial analyst or IT professional may be highly exposed because AI can perform an increasing number of tasks within those occupations. However, AI may make those workers more productive rather than eliminate their jobs. An accountant using AI to analyse information, investigate variances and prepare reports may simply accomplish considerably more work than an accountant without it. The more important question is therefore whether AI complements the worker or substitutes for the worker.
The ILO makes a similar distinction. Its research suggests that the broader effect of generative AI is more likely to involve the transformation or augmentation of jobs than wholesale occupational replacement. That distinction becomes critical when examining Barbados.
The IMF Has Already Identified The Risk
Research published by the International Monetary Fund in 2024 provides perhaps the strongest indication of Barbados’s potential exposure. The IMF divided employment into three broad categories based upon exposure to AI and the degree to which AI complements human labour.
Its estimates for Barbados were:
- 50% of the labour force — low AI exposure
- 29% — high AI exposure with low complementarity
- 21% — high AI exposure with high complementarity
That means approximately half of the Barbadian labour force is in occupations considered highly exposed to AI. However, the distinction between those latter two groups matters enormously. The 21% in highly exposed but highly complementary occupations could potentially experience substantial productivity gains as AI assists workers. The 29% classified as highly exposed with low complementarity is arguably the group deserving the greatest attention. These are jobs where workers have high exposure to AI but relatively less potential for AI simply to complement what they do.
This does not mean 29% of Barbadian jobs will disappear. It does mean that a substantial portion of the country’s labour market sits on the more vulnerable side of the AI exposure equation.
AI Does Not Need To Eliminate An Occupation To Affect Employment
Suppose a department currently requires ten employees to process information, prepare documents, answer routine enquiries and produce reports. If AI eventually allows the same volume of work to be performed by six employees, AI has not technically “replaced” the occupation. But four positions are no longer required. The effect could also occur gradually. Companies may simply stop replacing employees who resign or retire. Entry-level positions may decline. Existing employees may be expected to handle larger workloads using AI tools.
This means the impact may not initially appear as mass layoffs. Instead, it could emerge through slower hiring, fewer entry-level opportunities, changing job descriptions and higher productivity per employee. Traditional unemployment statistics alone may therefore be insufficient for understanding what is happening. We will also need to watch hiring, vacancies, wages, productivity and changes in employment within particular occupations.
Blue Collar Versus White Collar May Be The Wrong Divide
The emerging divide may therefore be less about blue-collar versus white-collar employment and more about the nature of the tasks performed. Jobs dominated by predictable digital tasks are increasingly exposed. Jobs requiring physical presence, dexterity, unpredictable physical environments, substantial human interaction or responsibility for complex decisions may prove more resilient — even when AI changes parts of those occupations.
For Barbados, that raises some uncomfortable questions. A highly educated, service-oriented economy contains substantial numbers of people whose daily work increasingly overlaps with the capabilities of artificial intelligence. At the same time, many trades and physically oriented occupations remain comparatively insulated from generative AI, at least for now.
Artificial intelligence therefore represents both an opportunity and a labour-market risk for Barbados. The challenge is not simply to ask whether AI will “take our jobs.” A much more useful question is:
“How much of the work Barbadians currently get paid to do will humans still need to do five or ten years from now — and what should we be doing today to prepare for that change?
Amit Uttamchandani is a Barbados-based technology executive with more than 20 years of experience across the Caribbean. He has a longstanding interest in using data to explore economic, social and technological issues affecting the region. He is the founder and editor of caribsignal.com. You can connect with him via LinkedIn.