Artificial intelligence could unlock up to $136 billion in productivity gains across Africa, but only if countries allow data to move securely and seamlessly across borders, according to Microsoft.
Speaking in an interview, Microsoft Africa’s Government Affairs Director, Akua Gyekye, said Africa’s limited computing infrastructure and restrictive data policies threaten the continent’s ability to scale AI adoption.
“Africa has barely 1–2 per cent of global compute capacity,” Gyekye said. “If we are serious about spreading AI across the continent, infrastructure matters—but so does the right policy environment.”
Infrastructure gaps and regional solutions
AI systems require massive computing power, yet Africa controls less than 1 per cent of global capacity. Although governments have urged global cloud providers to build local data centres, the continent had just 223 data centres across 38 countries as of mid-2025—less than 0.02 per cent of the global total.
Microsoft, which launched Africa’s first hyperscale cloud data centres in South Africa in 2019, operates regions in Johannesburg and Cape Town. However, Gyekye warned that a purely national approach may be inefficient.
“A regional model makes more sense,” she said. “If you build in Nigeria, how does the rest of West Africa benefit? If you build in South Africa, how do neighbours connect?”
This raises a critical policy question: can countries allow data to reside in neighbouring jurisdictions while still enforcing national data-protection laws?
Data sovereignty versus data silos
Africa has rapidly expanded data-protection regulation over the past decade. From just 12 countries with data laws in 2012, at least 46 now have legislation or frameworks, supported by the African Union Convention on Cyber Security and Personal Data Protection, which came into force in 2023.
Enforcement has intensified. In Nigeria, regulators fined Meta $220 million in 2024 following a joint action by the Federal Competition and Consumer Protection Commission and the Nigeria Data Protection Commission over alleged intrusive data practices.
While such measures aim to protect citizens, Gyekye cautioned that strict localisation rules can create “data silos” that slow innovation and regional expansion.
“Cybersecurity threats don’t respect borders,” she said. “In many cases, global cloud providers can offer stronger protection than local hosting.”
Sovereign cloud and AfCFTA ambitions
To balance sovereignty with scale, Microsoft is promoting a “sovereign cloud” model—where data may be hosted outside a country but remains subject to its laws through encryption, local key management and strict access controls.
The approach aligns with the goals of the African Continental Free Trade Area, which the African Union estimates could boost intra-African trade by over 50 per cent. Analysts say digital trade, fintech, logistics and AI-driven services under AfCFTA all depend on cross-border data flows.
“Without data collaboration, AfCFTA’s promise cannot be fully realised,” Gyekye said.
Beyond infrastructure: skills and inclusion
Microsoft says infrastructure alone will not deliver Africa’s AI dividend. Through its Airband initiative, the company claims to have connected more than 117 million Africans to broadband, while also investing in AI skills for students, developers and policymakers.
Language accessibility remains another hurdle. In February 2026, Microsoft unveiled expanded AI capabilities in 39 African languages through its Paza initiative, aimed at making AI tools more locally relevant.
“If AI doesn’t speak your language, you won’t use it,” Gyekye said.
The $136bn question
Analysts say AI could transform sectors such as agriculture, healthcare, education and logistics across Africa. But Gyekye warned that unchecked data localisation could fragment the continent’s digital economy, raising costs and slowing innovation.
“Data can flow while sovereignty remains protected,” she said. “That balance is how Africa unlocks AI’s full economic potential.”
If infrastructure, governance and skills align, Microsoft believes the projected $136 billion AI boost could be only the beginning.
