Most of the value in an African portfolio company is created after the funding round closes. Capital opens the door, but returns tend to come from the operational work that follows: better systems, sharper management and tools that let a lean team do more. Gene Grand regards artificial intelligence as the newest of those tools, and one that is far easier to talk about than to put to work.
“Every pitch deck I see now has AI somewhere in it,” he says. “Far fewer can tell me which process it has improved, and by how much.”
Gains without the same disruption
The World Bank’s 2026 assessment gives solid grounds for optimism. By its estimate, AI could make a meaningful difference to productivity in 16.2% of jobs across developing economies, not far below the 18.7% figure for high-income countries. The risk of displacement is far lower. Just 4.5% of existing jobs in low- and middle-income countries are at risk of automation by generative AI, against 14.2% in high-income countries.
For businesses in developing economies, African companies among them, that is an unusual balance for a new technology: much of the upside, with a fraction of the disruption.
The African Development Bank has put a figure on the potential. Its December 2025 report projects that inclusive AI deployment could generate up to US$1 trillion in additional GDP across Africa by 2035, with agriculture accounting for 20% of the gains.
Adopt, adapt, then advance
The World Development Report 2026 sets out a sequence rather than a leap. Countries first adopt available tools, then adapt them to local conditions, and only over time advance towards frontier AI development. Each step requires more investment, skills and infrastructure than the last, and the report describes building frontier models as by far the hardest and most expensive path, one that is not a realistic near-term goal for most developing countries.
That ordering suits portfolio companies well. Most will create more value by fitting existing tools to local customers, data and languages than by trying to build models of their own.
Foundations first
Adoption only delivers when the basics are in place. The World Bank names power, internet access, data, skills and institutions as the foundations needed to use AI effectively. The report also stresses the need for strong education systems and business environments that let AI builders raise capital and experiment.
Much of that groundwork is still being laid, and at scale: Mission 300 aims to provide energy access to 300 million people across the region by 2030.
“The companies that get the most from AI are usually the ones that had already done the dull work,” he says. “Clean data, dependable systems and people who know how to use them.”
Testing the claim
Research on frontier firms supports that view. A World Bank background paper on growth-oriented companies backed by the International Finance Corporation found that 40% had adopted AI in the 2025 reporting cycle. Digitalisation was the strongest correlate of adoption, and the authors concluded that “prior capability rather than resources alone is key to adoption”. Adoption was lowest among firms headquartered in Sub-Saharan Africa and the Middle East, which points to room for growth rather than a ceiling.
The same paper cites a study in Kenya of a GPT-4-based business assistant that found no average effect on revenues and profits. High performers increased revenue by more than 15%, while results for low performers fell by around 10%. The tool magnified whatever was already there.
For investors, that shapes the questions worth putting to any company claiming an AI advantage. Which specific process has improved, and what was measured before and after? Is the underlying data reliable? Do staff have the skills to use the tools well, and can power and connectivity support them day to day?
“AI won’t rescue a business that isn’t working,” Gene Grand says. “In a company that already runs well, it can make a good team considerably better.”
The early evidence suggests AI will widen the gap between well-run African companies and the rest. Investors who stay close to operations after the round closes are the ones most likely to know which side of that gap their companies sit on.
