Will the sub-Saharan African economy manage to accelerate its pace in 2026 while facing a succession of global shocks? Behind the nuanced response of the World Bank, one certainty emerges: without deep digital transformation and successful integration of IA into production processes, the current momentum risks stalling.
Indeed, sub-Saharan Africa is demonstrating robust economic resilience. According to the report "Africa’s Pulse: Setting the Foundations for IA", published in October 2026 by the World Bank Group, the regional PIB growth is expected to reach 4,3% in 2026, compared to 4,1% in 2025.
This growth is explained by the rise in global trade and the recovery of countries that export oil and minerals. In the region, better management of public funds has helped reduce government deficits to 0,5% of PIB in 2025, with the goal of rebalancing budgets by 2028.
However, this improvement is little felt in the daily lives of the inhabitants. Average income per capita is only increasing by 1,8% in 2026, which is too low to reduce poverty.
The extreme poverty rate remains very high: it moves from 47,8% in 2026 to 47,1% in 2027 (meaning less than 3 dollars per day per person). And as the population grows rapidly, the total number of people in extreme poverty continues to rise.
The burden of debt. While public debt stabilized at 57% of PIB in 2025, its repayment weighs heavily on public finances.
According to the World Bank, external debt service has absorbed between 1,6 and 1,7% of PIB since 2021. Furthermore, interest payments alone represent between 2,9 and 3,2 percentage points of public spending over the 2023-2026 period.
All of this has the consequence of depriving states of the financial means to invest. Added to this is a resurgence of price tensions: the average inflation rate moved from 3,7% in 2025 to 5,5% in 2026, mainly due to difficulties in energy and transport.
At the same time, international financial aid granted by wealthy countries is falling sharply, recording a drop of 16 to 28% in 2025. Finally, extreme climate disruptions like the El Niño phenomenon threaten harvests and food supplies for nearly 49 million more people worldwide.
Colossal potential of IA. Faced with these constraints, artificial intelligence (IA) presents itself as a transformative engine of unprecedented scope.
In a full-exploitation and highly ambitious scenario from the World Bank, IA could add up to 1000 billion USD to Africa's PIB by 2035. This value creation would allow for the creation of 35 to 40 million digital jobs while generating approximately 150 billion USD in annual tax revenues for public administrations.
Global demand for digital infrastructure is already supporting African countries that produce critical minerals such as copper, cobalt, nickel, manganese, or platinum. With this theoretical potential, the ideal scenario is enticing.
But the reality on the ground shows a glaring gap. Fragile foundations.
According to the report's authors, the continent's current level of readiness leads to much more modest estimates. The integration of IA is expected to add only 0,2 to 4% to sub-Saharan Africa's cumulative PIB over the next decade.
To explain this fragility, the Bank points to significant infrastructure gaps. More than 900 million Africans still lack access to the Internet.
The continent accounts for 18% of the world's population, but holds only 0,6% of the world's datacenter capacity. Furthermore, barely 5% of these facilities are ready to handle workloads related to IA.
Moreover, only 2% of the training data used worldwide comes from Africa, clearly illustrated by the fact that Swahili only has 710 datasets on the Hugging Face platform, compared to over 88 000 for English. Uneven adoption.
Another concern is that the adoption of IA remains very uneven. At the start of 2026, the use of generative IA among the workforce varied from 7,2% in Rwanda to 23,1% in South Africa, remaining below 10% in 16 countries.
In large companies, usage reaches 44% in Kenya and Nigeria, compared to 61% in the United States. In the public sector, usage remains basic: 90,8% of identified projects manage internal administrative tasks, 88,9% rely on simple predictive models, and only 1,4% leverage generative IA.
In the labor market, the direct exposure of jobs to automation remains low (2,6% of positions compared to 14,2% in wealthy countries). Conversely, the potential for support and complementarity reaches 15,2%, but it requires reliable Internet access