The four largest tech economies in Africa are currently developing comprehensive artificial intelligence strategies. These nations recognize their heavy reliance on major U.S. companies such as Google, Microsoft, Nvidia, and Meta for essential infrastructure, and they are seeking greater autonomy over the terms of these partnerships.
Countries like Nigeria, Egypt, and Kenya have issued draft AI policies highlighting the dependency on U.S. tech firms as a potential threat to national security. These concerns are expected to become increasingly pressing from January 2025 onward. Notably, South Africa reached a similar conclusion in a draft it published earlier this year, which had to be retracted due to inaccuracies generated by the AI tools used in its creation.
According to AI and policy experts consulting for African governments, many nations are largely dependent on American firms for computing power, investment, and technical expertise. As a result, there is a growing call for data sovereignty, development of local talent, and better terms from foreign technology providers.
Rachel Adams, founder of the Global Center for AI Governance, underscored that the push for digital sovereignty in Africa does not imply total independence from global AI supply chains. Instead, it reflects a desire for stronger controls over sensitive data, improved public procurement regulations, investments in local infrastructure, the creation of datasets in African languages, and clearer accountability from foreign AI providers.
Challenges of Big Tech Dependency
The World Economic Forum reports that although Africans comprise 18% of the global population, the continent’s data center capacity accounts for less than 1% of the total worldwide. Research from McKinsey reveals that the combined production capacity of Africa’s leading five markets is projected to fall short of France’s capacity by 2024.
In a bid to establish AI infrastructure, African firms are increasingly leveraging Western technologies. For instance, Cassava, led by Zimbabwean entrepreneur Strive Masiyiwa, partnered with Nvidia to inaugurate Africa’s first AI factory in South Africa in March. Additionally, iXAfrica, a data center service provider in East Africa, is collaborating with Oracle to establish Kenya’s first public cloud region.
However, significant challenges persist. Microsoft’s ambitious $1 billion partnership with G42 in Kenya has encountered hurdles after the government withheld a key commitment to purchase computing power. Furthermore, several open-source AI initiatives in Africa that have received funding from Meta are currently reliant on Google Cloud for their operations, according to Hilda Balasa, a senior policy advisor at the Tony Blair Institute for Global Change.
Kofi Yeboah, a researcher at the University of Cambridge, has expressed concerns about the impact of Western involvement in shaping Africa’s AI landscape. A partnership between Anthropic and Rwanda highlights this tension, as it raises questions about who ultimately sets the priorities for AI development in the region. Ayanthra Alayange from the Global Center for AI Governance pointed out that while opportunities for training Rwandans exist, significant ethical questions also arise regarding the long-term implications of such deals.
Efforts to Regain Autonomy
Some African nations are actively working to mitigate data leakage beyond their borders. Recently, Ghana, Nigeria, and Zambia rejected agreements to share health data with U.S. entities, opting instead to keep their citizens’ information secure within their territories.
Adeola Bojwoye, the Nigerian director of the Digital Impact Alliance, indicated that the lack of robust infrastructure has led African governments to consider “segmented” data processing setups. This model allows data to be processed outside the country while still stored domestically, emphasizing that ownership of infrastructure is merely the first step toward meaningful control.
However, experts caution that achieving actual autonomy is complex. Tay Pei-chin, a policy leader at the Tony Blair Institute, noted that while several North African countries have built data centers, many have outsourced their management to foreign firms, effectively compromising their control over this crucial resource. In her view, genuine control is about more than just ownership—it’s about maintaining operational integrity and security.
Future Initiatives in AI Development
In July 2024, the African Union rolled out its Continental AI Strategy, and by November 2025, the Smart Africa nonprofit will establish the African AI Council to facilitate resource sharing. Furthermore, the recently announced $60 billion Africa AI Fund aims to bolster infrastructure, foster talent, and support startups, including the allocation of 12,000 Nvidia graphics processing units to key centers in the Big Four and Morocco.
According to Priya Vora, CEO of the Digital Impact Alliance, there is a strong aspiration among African nations to operate as a cohesive digital market. However, this vision is tempered by concerns about reliance on both China and the United States, and the influence of their commercial interests.
While competition for foreign investment to become AI hubs continues to intensify, Adams notes that many projects, such as Nigeria’s Awari, keep their backend infrastructure opaque, complicating assessments of their dependence on Western technologies. Barrasa from the Tony Blair Institute advocates a regional approach, arguing that collaboration is essential, as no single country possesses the capacity to tackle these challenges in isolation. Trust among nations remains a significant barrier to this cooperation, making it likely that bilateral negotiations will continue to dominate the landscape.
