Region-Specific Context (RSC) in AI: Representation, Accuracy, and the Power to Shape Our Own Future
In a world increasingly shaped by artificial intelligence, the stories that are told, the solutions that are recommended, and the data that drives decisions often come from somewhere else. They come from cities and countries that are overrepresented in the training data of large language models like ChatGPT. But what about Jamaica? What about Trinidad, Guyana, Honduras, and the Dominican Republic? What about rural towns, small islands, and border communities that live rich, complex lives but exist only faintly, if at all, in these digital brains?
That is why I designed Region-Specific Context (RSC), a methodology that ensures people, problems, and potential from underserved regions are not just seen but empowered. RSC gives voice, accuracy, and relevance to those living and working outside the global AI spotlight.
The Problem We Are Solving
Large language models are trained on oceans of data from the internet, books, and media. But these oceans are uneven. Most of the content comes from a few countries, a few languages, and a few dominant cultures. This leads to several hidden but serious problems:
Lack of Representation
If your community is not in the data, your reality will not shape the answers. A child in St Lucia asking about farming tools may get advice for European weather. A Jamaican MSME looking for pricing strategies may get a Silicon Valley-style response that misses the mark.Poor Accuracy
LLMs make confident claims that may be wrong or irrelevant in your context. For example, suggesting digital payment solutions in areas where smartphones are scarce or assuming trust in institutions where mistrust is the norm.Perception Risks
Over time, people begin to believe that local knowledge is not valuable. Children and young adults may accept AI’s answers as truth, even when those answers erase their culture, speech patterns, or lived experience.Societal Harm
If local businesses make decisions based on generic, globalized AI outputs, they risk failure. If governments use AI to draft policy without RSC, they miss critical nuances. If education systems adopt AI tools that do not speak our language, literally and culturally, young minds are shaped by invisible bias.
Why We Need RSC
Region-Specific Context is about balance and fairness. It does not fight against AI; it upgrades it. It ensures that AI supports people, rather than replacing or misrepresenting them.
RSC is particularly vital for:
Children, who deserve learning tools that speak to their reality
Youth and adults, who need job tools that understand local constraints
MSMEs, who cannot afford to waste time and money on wrong-fit advice
Corporate leaders, who must make informed decisions that respect cultural dynamics
Policymakers, who need to protect citizens from invisible algorithmic harms
The Benefits of RSC
Cultural Relevance: Advice, prompts, and strategies reflect your lived experience
Economic Alignment: Pricing, purchasing, and consumer behavior are matched to regional norms
Empowerment: Users inject their own data and insight, becoming co-designers of their solutions
Increased Accuracy: The AI is corrected, not just queried
Fairness in AI Adoption: Communities gain equitable tools, not hand-me-down algorithms
The General Approach
RSC works by teaching people to add local insight into their AI prompts. It does not require coding or technical skill. It requires knowing your environment and using that knowledge intentionally.
We do this by:
Inserting local behaviors, challenges, and preferences directly into prompts
Stating your location and its realities, such as infrastructure gaps or holiday seasons
Describing trust dynamics, such as reliance on face-to-face transactions or community leaders
Specifying economic and digital conditions, like access to credit or use of WhatsApp over email
For example:
Instead of: “Generate marketing ideas for a small restaurant”
Try: “Generate marketing ideas for a small restaurant in rural Belize, where customers rely on cash, radio ads, and weekend markets”Instead of: “Create a business growth strategy”
Try: “Create a business growth strategy for a Trinidadian MSME that sells beauty products through Instagram and courier delivery, but faces customer trust issues with pre-payment”
The RSC Framework
The framework is simple, structured, and designed for everyone:
Add Geographic Context
Name your location, from the country to the specific town or neighborhoodSpecify Economic Realities
Mention informal markets, credit norms, purchasing power, or delivery challengesInclude Cultural Nuance
Share language quirks, local holidays, trust barriers, or social normsInject Real-World Observation
Add what you actually see or experience, such as “most people pay in cash on Friday after payday”State Your Role or Limitations
Let the AI know if you are a single parent, a teacher with limited tech, or a youth leader in a low-connectivity area
Final Thoughts
Artificial intelligence should not be a one-size-fits-all engine of global advice. It should be a conversation, where your voice matters. The future belongs to those who can train the trainer, who can guide the AI, and who can insert their reality into the machine’s imagination. That is what Region-Specific Context allows. We are not just teaching the AI about ourselves. We are reclaiming the right to be seen, heard, and served. This is your region. This is your voice. Let it shape the future.
About the Author
Adrian Dunkley is a 15-year A.I scientist, entrepreneur, and founder of StarApple AI, the first AI startup in the Caribbean. He has built more than 150 AI solutions across finance, insurance, tourism, and education. As a recognized AI thought leader and advisor to governments and businesses, Adrian designed the Region-Specific Context (RSC) methodology to ensure that underrepresented communities are empowered, not erased, by artificial intelligence. His work combines AI, behavioral science, and regional equity to help shape a smarter, fairer future.
