Where Is Africa’s Data Actually Being Stored?

When a banking transaction is processed in Lagos or a payroll system runs in Cape Town, the data often leaves the continent before the task is complete. Instead of remaining within African infrastructure, it is routed through subsea cable systems to the global internet’s established hubs: Frankfurt, London, Amsterdam, Paris, Dublin, and increasingly Northern Virginia. 

This is the contradiction at the center of Africa’s digital sovereignty ambitions. Over the past decade, more than 40 African countries have introduced data protection and localization frameworks. Yet most of the continent’s data still lives offshore. Estimates from institutions such as the IFC and the GSMA suggest that more than 80% of African data  is hosted outside Africa. 

The issue is not a lack of legislative ambition. It is the absence of enough local infrastructure capable of supporting modern digital economies at scale. 

In this edition of #TechTalkThursday, we examine why Africa’s data still lives abroad, what is slowing the shift toward local hosting, and why the rise of AI is making the infrastructure conversation more urgent than ever. 

The Invisible Path of African Data 

For decades, Europe’s major internet hubs became the default destinations for global hosting because they offered what enterprises needed most: reliable power, dense connectivity, mature ecosystems, and operational stability. African organizations adopted those locations for practical reasons. 

For many enterprises, hosting in Frankfurt or Dublin has historically meant higher uptime guarantees, better redundancy, lower latency across international systems, and easier access to hyperscale cloud platforms. Local alternatives often struggled with power instability, limited rack density, cooling challenges, and financing constraints. 

The result is a widening gap between regulation and operational reality. Countries such as Nigeria  and Kenya  have strengthened data governance and localization policies, but laws alone cannot create hyperscale infrastructure. 

The capacity gap illustrates the scale of the challenge. Africa’s estimated demand for data center infrastructure is approaching 1,000 megawatts, while active installed capacity remains closer to 360 megawatts. Nearly half of that capacity sits in South Africa alone. 

For much of the continent, meaningful local hosting options still do not exist. That leaves governments in a difficult position: mandating local data residency while lacking the infrastructure required to support it. 

At that point, sovereignty becomes difficult to enforce physically, even if it exists legally. 

Solving the “Three-Body Problem”  

The infrastructure conversation often focuses on servers and cables, but the real constraint is power. 

Hlumelo Fungile, Chief Commercial Officer at PAIX Data Centres, frames this gap as a “three-body problem” a misalignment between capital, operators, and hyperscalers. In his 2026 analysis of the sector , he noted that infrastructure requires a rare synchronization of forces where physics does not negotiate: 

“Energy moves at the speed of turbines and transmission lines. Capital moves at the speed of risk repricing. Hyperscalers move at the speed of software. Where those speeds synchronize, infrastructure gravity forms.”  

— Hlumelo Fungile, CCO, PAIX Data Centres 

Energy infrastructure develops slowly through transmission projects, generation capacity, and national grid expansion. Capital responds to perceived risk and macroeconomic stability. Hyperscalers move faster, scaling wherever infrastructure already exists. 

Where those forces align, large-scale digital infrastructure emerges. Where they do not, investment stalls and data continues flowing offshore. 

That is why digital sovereignty is increasingly becoming an energy conversation. Governments can establish regulatory frameworks, but without stable power systems and predictable infrastructure policy, hyperscale investment remains limited. 

The challenge becomes even more visible in AI infrastructure planning. AI workloads require significantly higher power density than traditional cloud hosting. In markets already dealing with energy deficits, this creates difficult trade-offs. 

In Nigeria, for example, a single 50 MW hyperscale campus could consume close to 1% of the country’s peak generation capacity. That figure illustrates how closely Africa’s digital ambitions are now tied to national energy strategy. 

Why Regional Trust Matters More Than Ever 

Building fully independent hyperscale ecosystems across all 54 African countries is economically unrealistic. As a result, industry leaders increasingly support a regional hub model. 

Under this approach, cities such as Nairobi, Lagos, Johannesburg, and Casablanca would function as regional digital infrastructure anchors serving neighboring markets. The economics are compelling: larger regional facilities reduce deployment costs, improve utilization, and attract stronger cloud ecosystems. 

The obstacle is political trust. 

Governments want assurances that sensitive systems such as health records, tax infrastructure, and digital identity platforms remain protected and accessible even when hosted across borders. 

That is where new regional frameworks are beginning to emerge. African Union initiatives, regional economic communities, and pan-African investment structures are pushing toward harmonized standards for data governance, identity systems, and cross-border digital operations. 

At the same time, stronger regional peering infrastructure is reducing dependence on international traffic routing. Instead of African internet traffic traveling to Europe before returning to another African country, more traffic can now remain within regional exchange systems. 

The success of this model will depend on enforceable governance frameworks, transparent procurement structures, and long-term regulatory stability. Without trust, regional infrastructure cannot scale. 

Why the AI Era Raises the Stakes 

The data sovereignty conversation was already urgent, but the arrival of AI has transformed it from a legal nice-to-have into a functional necessity. This shift is happening on two fronts: the physical design of the data center and the “sovereign logic” of the models themselves. 

On the physical layer, AI is disrupting the infrastructure by moving the bottleneck from physical space to power density. This is not an incremental upgrade; it is a fundamental shift in how facilities must be designed. In a market like Nigeria, this power requirement is a critical constraint. To put the scale in perspective, a single 50 MW hyperscale campus would consume nearly 1% of Nigeria’s entire peak generation capacity. It highlights a stark reality: Nigeria is not waiting for digital demand; it is waiting for an energy infrastructure capable of sustaining the “intelligence layer” of a modern economy. 

Beyond the megawatts, there is the question of utility. AI systems are only as useful as the data they are built on. If African data continues to be stored and processed offshore, the AI tools built for the continent will remain “foreign” in their logic. 

The practical consequences are significant. A diagnostic tool trained on European health records may miss biological patterns specific to sub-Saharan populations. Similarly, a crop monitoring system built on foreign datasets will not reliably identify diseases affecting cassava or sorghum in East Africa. For AI to be locally relevant and effective, the data must stay within African infrastructure, allowing models to be trained, refined, and improved in the context where they are applied. We are no longer just building for storage; we are building for the ability to process African reality locally.  

The Window Between Now and 2030 

Africa’s digital infrastructure momentum is growing. Subsea cable systems such as Google’s Equiano and Meta’s 2Africa are improving continental connectivity. Investment activity is increasing across key markets. Data center expansion continues in West, East, and Southern Africa. 

But the financing gap remains substantial. Africa currently attracts only a fraction of the digital infrastructure investment required to meet projected demand by 2030. Estimates from firms such as McKinsey suggest the continent may require between $20 billion and $25 billion in infrastructure investment this decade to close the gap. 

The next few years will determine whether Africa’s digital sovereignty ambitions become operational reality or remain policy aspirations. 

Because ultimately, sovereignty is not defined by legislation alone. It is defined by whether the infrastructure exists to support it.