4 June 2026
Why “just use ChatGPT” is quietly failing Caribbean businesses
Use case — A marketing team that “uses AI”
Walk into almost any office in Bridgetown, Port of Spain or Kingston today and you’ll find the same thing: people already have AI open in a browser tab. ChatGPT, Gemini, Copilot — access is no longer the problem. And yet the work isn’t getting noticeably better. Reports still take as long. Campaigns still feel generic. The promise hasn’t landed. That’s because access and capability are not the same thing. Opening ChatGPT is like being handed the keys to a commercial kitchen. The equipment is world-class — but if nobody taught you to cook, you’ll still make the same sandwich you always made, just faster and with more anxiety about whether you’re doing it right. Take a regional FMCG marketing team — a setup I’ve seen across the categories I worked in for over a decade. Everyone on the team “uses AI.” But dig in and you find each person is pasting the same vague prompt (“write me a social caption”), getting bland output, lightly editing it, and shipping it. The AI hasn’t changed how they think about the work. It’s just become a faster way to produce mediocre. The teams getting real leverage do something different. They use AI to decide what shouldn’t be made at all — to pressure-test whether a campaign idea is differentiated before a dollar is spent, to summarise three competitor launches in minutes, to draft five strategic angles so the human can pick the sharpest. The tool is the same. The judgment around it is the entire difference.
Tool — Prompt structure, not prompt tricks
Forget the “100 magic prompts” lists. The single most useful habit is giving the AI three things every time: context (who you are and what the business is), the specific task, and the constraints (length, tone, what to avoid). “Write a caption” gets you sludge. “You’re writing for a Barbadian rum brand targeting 25–40s; give me three Instagram captions under 15 words, confident not corny, no exclamation marks” gets you something usable. The skill isn’t the tool. It’s the instruction.
Regional story — The Caribbean’s quiet advantage
There’s a real advantage hiding here for the Caribbean. We don’t carry decades of legacy AI infrastructure or entrenched bad habits to unwind. A small business in St. Lucia or a marketing team in Barbados can adopt current best practices from the start — if someone shows them what those practices actually are. The gap isn’t talent or ambition. It’s structured capability-building, delivered with regional context instead of recycled Silicon Valley headlines.
Tip — Turn one task into a system this week
Pick one task you do every week — a status update, a recurring report, a set of captions. Write one reusable prompt for it that includes your context, the task, and your constraints. Save it. You’ve just turned a one-off into a system. Do that ten times and you’ve quietly rebuilt how your week runs.
Want your team to actually build this capability, not just talk about it?
Join the Amplifi AI Masterclass →FAQ
Frequently asked
- Do I need technical or coding skills to learn AI for business?
- No. Practical AI capability is about judgment and clear instruction, not code. Most business value comes from knowing where AI fits in your workflow and how to direct it well — skills any professional can build.
- Is AI training worth it if my team already uses ChatGPT?
- Often it’s exactly those teams that benefit most. Access without training usually means people are using a fraction of the capability and reinforcing old habits faster. Structured training closes the gap between using AI and using it well.