Closing the Digital Gender Divide Before AI Makes It Permanent

Every time a Caribbean government official speaks about the digital gender divide, the conversation gravitates toward coverage maps and mobile penetration rates. More towers. Better broadband. Subsidised data plans. These are real needs, and they matter. But they are not the fundamental problem, and treating them as if they are has allowed a governance failure to persist under the cover of an infrastructure narrative.

My argument is this. The Caribbean digital gender divide is not mainly a connectivity problem. It is a governance failure, and without deliberate, structured policy action, the AI systems being adopted across the region's economy will lock these inequalities into the economic infrastructure for a generation. Once those systems are trained, deployed, and normalised, the data gaps that produced their biases disappear from view. The discrimination becomes architectural.

What the Data Actually Shows

The GSMA Connected Women Commitment initiative and its associated annual reporting have tracked the mobile gender gap across developing regions for years. The GSMA Connected Women Report 2024 confirms that women in low- and middle-income countries remain far less likely than men to own a mobile phone and to use mobile internet. The gap has narrowed in some markets and held firm in others.

The barriers the GSMA data consistently identifies are not primarily about infrastructure. They are about affordability, literacy, social norms around women's technology use, safety and harassment online, and the perceived relevance of digital content to women's lives. These are social and policy problems. They respond to social and policy interventions: subsidised devices targeted at women, digital literacy programmes designed for women's contexts, regulatory action against online harassment, and content development that speaks to women's economic and social priorities.

The Intersecting Barriers Caribbean Women Face

Caribbean women do not face a single barrier to digital inclusion; they face a constellation of reinforcing disadvantages. Gender intersects with race, class, geography, age, and disability to produce patterns of exclusion that vary across the region and within individual countries.

Affordability is a consistent factor. Mobile data plans consume a higher proportion of income for women than for men, because women's incomes are lower on average. The gender pay gap is documented across CARICOM labour markets by the ILO and by national statistics offices.

Online safety is a second consistent factor that infrastructure investment does not address. Women across the Caribbean report experiences of harassment, surveillance, and abuse in online spaces. With no broad online safety legislation in most CARICOM jurisdictions, the internet is a more hostile place for women than for men.

At tertiary level, Caribbean women's educational attainment runs well ahead of men's. The claim that women lack the capacity to use digital technology does not survive contact with the data. What they lack is safe, affordable, relevant access.

How AI Converts a Gap into a Permanent Structure

AI systems learn from data. The quality of their outputs for any given population depends on the quality and quantity of data that population has contributed to training. Women who are less digitally connected produce less data. Less data means poorer AI performance for those populations.

AI-driven financial services will be less accurate for women who have thinner digital transaction histories. AI-assisted healthcare tools will perform worse for women whose health experiences are underrepresented in training datasets. Agricultural AI advisory services will be less useful for female farmers. Government service delivery AI will make more errors for women.

Each of these accuracy gaps, when embedded in production AI systems, becomes a self-reinforcing dynamic. Women receive worse service, trust AI systems less, engage with them less, contribute even less data, and the next generation of AI systems is trained on an even more gender-skewed dataset. The gap widens with each iteration.

The Governance Failures That Created This Problem

Digital inclusion policy in the Caribbean has been characterised by aspirational frameworks without implementation mechanisms, gender commitments without gender-disaggregated targets, and technology investment without equity analysis.

The CARICOM Regional Digital Economy Policy Framework establishes a regional vision for inclusive digital economies. But the Framework has not been operationalised with specific gender equity targets, funded implementation plans, or accountability mechanisms for member states.

Jamaica's Vision 2030 includes commitments to gender equality and to digital economy development. These commitments exist in different sections of the plan and have not been integrated.

Recommendations

1. Establish gender-disaggregated digital inclusion targets in all national digital economy plans. Every CARICOM member state with a digital economy strategy should add specific, measurable targets for closing the gender digital divide within defined timeframes.
2. Require gender impact assessments before deploying AI systems in public services. Any government ministry or public authority deploying an AI system for service delivery should be required to conduct and publish a gender impact assessment before deployment.
3. Fund Caribbean-specific digital safety legislation addressing online harassment. Model legislation developed at the CARICOM level, with member state implementation supported by technical assistance, would be the most efficient approach.
4. Integrate gender equity analysis into the CARICOM Regional Digital Economy Policy Framework's implementation structures. A dedicated technical working group on gender and the digital economy, with UN Women Caribbean as a partner, should oversee this process.
5. Invest in Caribbean-specific data collection on women's digital access and use. The Statistical Institute of Jamaica and its CARICOM equivalents should incorporate mandatory gender-disaggregated digital access and use questions into regular household and labour force surveys.
6. Create a CARICOM AI equity watchdog with gender expertise. A regional body with authority to review AI deployments in CARICOM member states for equity impacts, including gender impacts, would provide the accountability mechanism the current framework lacks.

Conclusion

The digital gender divide in the Caribbean is not waiting for a policy response. It is producing consequences every day. The AI transition underway across the region's economy will either widen or narrow these existing inequalities, depending entirely on whether governance frameworks require equitable design and equitable outcomes.

The window for preventing AI from making the digital gender divide permanent is measured in years, not decades. Governance must act before the architecture sets.

About the Author

Dr. Shirley Budall is a Caribbean expert in gender, inclusion, and AI governance with demonstrated experience in the ethical, legal, social and governance dimensions of artificial intelligence and digital technologies. Contact: insights@starapple.ai