advertisement
Who Is Financing Africa’s AI Ambitions?
Africa’s ambition to become a global player in artificial intelligence will require significant investment in computing infrastructure, data centres and AI development. However, while governments have announced billions of dollars in funding commitments, much of the investment currently building the continent’s AI capacity is coming from private companies and international technology firms.
When Darlington Akogo travelled from Accra to Kigali in April 2025 for the first Global AI Summit on Africa, he identified computing capacity as one of the continent’s biggest obstacles to AI development. The founder of minoHealth AI Labs, which develops medical AI solutions, told African Business that “the computing issue was identified as the biggest challenge for Africa.”
The summit produced an ambitious response. More than 50 countries signed the Africa Declaration on Artificial Intelligence, which included plans for a $60 billion Africa AI Fund intended to mobilise public, private and philanthropic investment.
advertisement
However, 19 months later, the proposed fund remains under development. A year after the Kigali summit, Smart Africa director general Lacina Koné told Rwanda’s New Times that the initiative would operate as a blended financing mechanism involving governments, private investors, venture capital firms, mobile operators and development partners. At the time, he said his team was still “crafting what the fund should look like”, with no funding commitments or money raised reported.
The delay comes as the cost of developing Africa’s AI infrastructure continues to rise. McKinsey estimates that demand for data centre capacity across the continent could increase from approximately 0.4 gigawatts to between 1.5 and 2.2 gigawatts by 2030. Meeting this demand would require between $10 billion and $20 billion in additional investment, excluding the computing equipment needed inside the facilities.
The cost of that equipment is substantial. Cassava Technologies’ plans to deploy 12,000 Nvidia GPUs across five African countries illustrate the scale of investment required. With individual chips estimated to cost between $45,000 and $60,000, the processors alone could account for between $540 million and $720 million.
advertisement
So far, investment is coming primarily from three sources: African private companies, development finance institutions and international technology firms.
Among African investors, Cassava Technologies, founded by Zimbabwean entrepreneur Strive Masiyiwa, has emerged as a major player. Its AI factory programme, which Bloomberg reported could cost up to $720 million, began with plans to install 3,000 GPUs in South Africa.
Group chief executive Hardy Pemhiwa has emphasised the importance of African companies taking the initial investment risk, saying that “we can’t expect others to go first.”
advertisement
Development finance institutions are providing another source of capital, although much of this funding is structured through loans and targeted investment programmes. The International Finance Corporation provided a $100 million loan to data centre operator Raxio Group in 2025 to support infrastructure development across several African markets, including Ethiopia and Angola.
In May 2026, the Africa Finance Corporation approved $100 million for Africa-focused technology fund managers. Meanwhile, the African Development Bank and the United Nations Development Programme launched the AI 10 Billion Initiative in February, targeting the mobilisation of up to $10 billion by 2035. Its initial phase includes a 10-month roadshow to engage governments and potential investors.
International technology companies, however, remain among the largest financiers of infrastructure expansion on the continent. Microsoft has committed R5.4 billion, approximately $300 million, towards expanding its South African cloud and AI infrastructure by the end of 2027. This is in addition to R20.4 billion previously invested in data centres in Johannesburg and Cape Town.
While such investments increase the computing capacity available in Africa, they also raise questions about ownership. Infrastructure may be physically located on the continent, but the companies financing and operating it often retain control of the underlying assets.
The disparity becomes clearer when compared with global technology spending. Amazon, Alphabet, Meta and Microsoft are expected to spend approximately $725 billion collectively in 2026, largely on AI infrastructure and data centres. That is more than 12 times the proposed $60 billion Africa AI Fund target.
For African governments, the challenge is therefore not simply attracting investment, but determining the terms under which that investment supports long-term digital sovereignty.
Kenya’s proposed $1 billion geothermal-powered data centre project involving G42 and Microsoft illustrates the complexity. The project encountered difficulties after investors sought government guarantees that, according to Invest Kenya’s Sandy Okoth, would have transferred significant risks to taxpayers. The concerns highlighted the tension between attracting much-needed infrastructure investment and protecting public resources.
For CIOs and enterprise technology leaders, the financing gap has direct implications for the cost and availability of AI services. Limited local GPU capacity means that running AI workloads within African data centres can be more expensive than using infrastructure in established global markets. Meanwhile, much of the new capacity depends on foreign investment or debt financing that requires commercial returns.
As more organisations deploy AI applications, their infrastructure decisions will influence where demand, investment and ultimately value accrue.
The Africa AI Council, established to advance the Kigali declaration, now has 15 members and has identified wider access to computing power as an initial priority. Its approach includes exploring shared GPU infrastructure that allows countries to access computing resources without independently financing and building their own facilities.
However, the question of how that infrastructure will be financed remains unresolved. The proposed $60 billion fund represents an ambitious commitment, but turning it into operational computing capacity will require actual capital, clear financing structures and sustained participation from governments and private investors.
For now, African companies and international technology firms are doing much of the heavy lifting. Whether African governments can translate their funding pledges into investments that give the continent greater ownership of its AI infrastructure remains one of the defining questions surrounding its digital ambitions.