Our insights in education, health, decentralized policies and human capital
Governments across Africa are being urged to embrace artificial intelligence. AI promises faster diagnoses, better-targeted social programs, smarter school planning and more responsive local governments. But every one of these promises rests on a single, often overlooked foundation: the quality of public-sector data. AI does not correct weak data — it learns from it, and then scales its blind spots.
ALG is launching a multi-country research initiative to answer a question that has become urgent: is Africa’s public-sector data ready for the age of AI — and what would it take to get there?
Why this matters now
Over the past two decades, African administrations have built impressive data systems: education management information systems, health information platforms, civil registration, local budgeting and public payroll databases. Development partners have invested heavily in them. Yet their reliability varies widely — across countries, across sectors, and sometimes across districts of the same country.
As governments and partners move toward AI-enabled services, this variation becomes a strategic risk. Decisions about where to build a school, how to allocate health staff or which households receive support may soon be shaped by algorithms trained on data whose limits are poorly understood. Understanding those limits is no longer a technical detail. It is a question of public trust and good governance.
Our vantage point
ALG has worked for more than twenty years across 22 African countries, conducting evaluations, nationally representative surveys, institutional assessments and capacity-building programs for governments and development partners. This long engagement gives us an unusual view of how public data is actually produced, managed and used — well beyond what official reports describe.
From this repertoire of 22 countries, the study will focus on a sample of ten, chosen to reflect the diversity of the continent: francophone and anglophone administrations, centralized and decentralized systems, and different stages of digital maturity.
What we will examine
The research concentrates on four areas where data shapes public decisions most directly:
- Education — how learning and enrolment data inform planning and resource allocation
- Health — how service and population data guide programs and staffing
- Decentralized policies — how local governments produce and use data for planning and accountability
- Human capital — how public institutions track and develop the skills of their workforce
Across these four areas, three questions guide our work: How reliable is the data? How is it actually used in decision-making? And what would it take to make it AI-ready?
How we will work
The study follows a comparative, multi-country design, combining a review of existing data systems, field inquiry, and structured dialogue with the institutions that produce and use public data. We will work in partnership with national counterparts — ministries, statistical offices and local authorities — and share findings with them first.
What to expect
The initiative will produce country briefs, a comparative synthesis across the ten countries, and practical recommendations for governments and development partners preparing AI-enabled public services. Our aim is not to rank countries, but to identify what works, what does not, and where investment in data quality will yield the greatest returns.
Get involved
We welcome conversations with ministries, national statistical offices, development partners, foundations and research institutions interested in contributing to, supporting or learning from this work.
Contact us at [email protected] or through our contact page.