Start at Five: What a US$1 Million Caribbean AI Bootcamp for Children Proved in One Month

Over 200 Caribbean young people joined The Genius Project in 2026. The youngest was five. The oldest was eighteen. Across one month they moved from using AI tools to building machine learning models, US$1 million in cash and prizes was awarded, and no family paid tuition. CAIA was one of the sponsors, providing regional backing and reach across CARICOM.

Regional AI policy currently concentrates on infrastructure, national strategies and workforce reskilling for adults already in employment. All three matter. None of them reaches the group that will hold these jobs in 2035, and that group is in primary and secondary school right now. This programme is the closest thing the region has to a working model for reaching them.

The 2026 Programme in Numbers

• Over 200 participants aged 5 to 18, from across the Caribbean, in person in Jamaica and virtually elsewhere.
• US$1 million in cash and prizes awarded across a single month, funded entirely by sponsors.
• Zero tuition charged to any family.
• Four problem domains: crime and community safety, poverty and access, sport, and AI ethics.
• A parallel parent track on safe AI use, critical thinking and misinformation.
• Roughly 15 percent completion across all programme areas, published openly by the programme.

The Curriculum Went Past the Tools

Most youth AI programmes stop at tool use. A student learns to prompt a chatbot, produces something that looks impressive, and leaves believing they understand AI. They understand an interface, which is a different thing and a far less durable one.

The Genius Project treats tools as the entry point and then moves the ground. Students went from prompting a model to understanding what a model is: training data, features, labels, and the difference between a system that learned a pattern and one that memorised an answer sheet. That requires statistics, and statistics requires mathematics. Distributions, probability, measurement error, and the reason a single result tells you almost nothing. For the older cohorts it meant Python, notebooks, and the ordinary experience of code that runs correctly and returns the wrong number.

Two subjects ran through every session regardless of age. Teamwork, since nothing was built alone. And problem definition, which is harder than it sounds and is the skill most adults in technical careers are still short of. A team that cannot state clearly what problem it is solving cannot build anything worth judging.

What the Teams Built

Crime and community safety. Teams examined where incidents cluster, how gaps in reporting distort what the data appears to say, and what a model can responsibly claim about a place or a person. Several groups worked out for themselves that a predictive model built on incomplete crime data largely predicts where the reporting is. Police services in far wealthier jurisdictions have paid consultancies substantial fees to arrive at the same finding later.

Poverty and access. Household budgeting tools, food price tracking, and matching people to services they qualify for but do not know exist. The constraint these teams hit, that Caribbean household data is thin and scattered, is a regional data problem CAIA has raised repeatedly. Thirteen-year-olds found it in a week.

Sport. Football and track data proved the strongest on-ramp to machine learning available. Students already had domain intuition, which meant they could immediately tell when a model was producing nonsense. Most beginners cannot, and that inability is precisely how bad models get deployed.

Ethics and responsible AI. Every team had to state who their system could fail, what data it should never hold, and what they would say to a person the model got wrong. This was a build requirement rather than a lecture module. Watching a fourteen-year-old explain why her model should not be used for hiring decisions is a stronger argument for AI ethics education than most policy documents manage.

Across all four areas the students built actual machine learning models. Systems that took input, produced output, and could be demonstrated to be wrong. That last property is the one professional practice depends on.

The Parent Track Is the Part Policy Should Copy

A child who understands AI better than every adult in the household is not in a safe arrangement. Parents ran their own track alongside their children, and this is the element CAIA considers most directly transferable to national programmes.

It covered the practical safety layer first: account and privacy settings, what a chatbot retains, what should never be pasted into one, and how to recognise a website built to harvest information. Then it moved to judgement. Telling a generated image from a photograph. Checking a claim before forwarding it. Recognising AI slop, the fluent and confident text that happens to be wrong, and understanding that the fluency is the trap rather than the reassurance.

The bar the programme set was modest and specific: a parent should be able to sit beside their child, look at the work, and ask one question that improves it. Regional misinformation campaigns aimed at the general adult population have struggled for years to achieve less than that. Reaching adults through their children turns out to work, because the motivation is already present.

The Completion Figure, Published

Completion across all programme areas currently stands at roughly 15 percent. The programme publishes it rather than reporting only enrolment, which is the norm in this sector and a habit that has made regional education data close to useless for comparison.

The programme identifies two drop-off points. The first is the transition from tools to mathematics, which is where any technical curriculum loses people. The second is infrastructure: unreliable connectivity and nowhere quiet to work. The second is a regional policy problem, not a curriculum problem, and it is the clearest illustration available of how connectivity gaps translate directly into lost technical capacity.

The Hackathon

The month closed with a final hackathon where teams presented to judges, defended their builds, and answered for their design choices. Congratulations to the winners, and to every team that presented at all. Defending a technical build in front of a panel of adults is difficult at thirty. A number of these presenters were not yet thirteen.

What CAIA Takes From This

CAIA draws four working conclusions. First, national AI literacy strategies across CARICOM should include a primary and secondary school component with a defined starting age, not a general commitment to youth engagement. Second, parent-facing AI safety education delivered through children reaches adults that standalone public campaigns do not. Third, connectivity is an AI skills issue, and the drop-off data from this programme is direct evidence of it. Fourth, programmes receiving public or donor funding should publish completion rather than enrolment, because the region cannot improve what it does not measure honestly.

The argument for starting at five is not that five-year-olds should write code. They should not. A five-year-old sorts, counts, spots patterns, and learns that a machine's guess can be wrong. That last lesson is the foundation of every good habit that follows. A child who learns it at five becomes a teenager who verifies and an adult who does not forward the deepfake. There is no cheaper intervention available to Caribbean governments, and no faster one.

Who Funded the 2026 Programme

The Genius Project charges families nothing, which works only because organisations across the region contributed cash and in-kind support. The 2026 sponsors were StarApple AI (prize funding, instructors and curriculum), Maestro AI Labs (technical mentorship and lab time), the Caribbean AI Association (regional backing and CARICOM reach), 14West (hackathon and prize pool), AI Trinidad and Tobago (delivery across the twin islands), Orbital Brand Science (in-kind support and family outreach), and Adrian Dunkley personally.

Families can register for the next cohort at beagenius.org. Tuition-free, from age five, virtual for participants outside Jamaica, no prior coding experience required.

Disclosure: Adrian Dunkley founded the Caribbean AI Association in 2024 and also founded The Genius Project, the programme reported here. CAIA states that connection plainly and reports the figures as the programme presents them.