That scale should make African governments, universities, investors and students ask a harder question: are we preparing only to use artificial intelligence, or are we preparing to build the foundations that make it possible?

Training people to use AI is important. It is the minimum entry ticket. But the global race is also about electricity, connectivity, computing power, data centres, local data, research capacity and institutions capable of turning knowledge into products.

The World Bank describes these foundations as the “four Cs”: connectivity, compute, context and competency. In simple terms, countries need reliable infrastructure, processing capacity, relevant data and skilled people. If one of those foundations is missing, ambitious AI projects become difficult to scale.

Rwanda already understands part of this challenge. Its National AI Policy says the country wants to become a global centre for AI research and innovation, while recent government initiatives have highlighted infrastructure, data systems, skills and governance as foundations of that ambition.

The opportunity now is to move from ambition to strategic positioning.

Rwanda and other African countries cannot outspend the world’s largest technology firms. 

They do not need to. Africa’s advantage can come from choosing where it can become indispensable.

One area is research. Universities should produce more advanced researchers, PhDs and technical teams who understand African problems deeply enough to create local solutions.

Another is talent. Young people need to stop seeing AI only as a threat to jobs and start understanding how work itself is changing. Some roles will shrink. Other roles that barely existed a few years ago will grow.

The third opportunity is investment attraction. AI infrastructure needs land, reliable energy, connectivity, security, regulation and predictable institutions. Countries that make these conditions attractive can host data centres, research operations, cloud infrastructure and specialised technology businesses even if they do not own the world’s largest AI companies.

Regional cooperation also matters. A single African market may struggle to justify certain infrastructure investments, while several connected markets can create scale. Shared research, interoperable rules and cross-border digital infrastructure can make the continent more attractive to serious investors.

For students, the message is equally clear. Do not prepare for the labour market you grew up hearing about. Prepare for the one being built now. Learn the tools, but also learn mathematics, data, engineering, research, business and the domain in which you want to solve problems.

For governments, the lesson from Meta’s spending is not to imitate Meta’s budget. It is to understand what that budget is buying.

Africa’s dangerous future is not a future without AI. It is a future in which Africans use the world’s most powerful AI every day while owning little of the research, infrastructure, data value or intellectual property behind it.

The goal should therefore be bigger than AI adoption. Africa must build enough capability to become a place where the next generation of AI is researched, hosted, adapted and created.