Nvidia Backs African AI Data Centre Rollout with Cassava
Nvidia has partnered with Cassava Technologies (Cassava) in a US$700m deal to develop AI-ready data centres across multiple African countries.
The project, first announced in March, marks Nvidia’s first direct infrastructure deployment on the continent and aims to provide access to its advanced graphics processing units (GPUs) in markets that have historically lacked sufficient compute capacity.
The collaboration contrasts with state-led Chinese investments in Africa’s digital infrastructure. Instead, Nvidia’s approach is based on private-sector execution, with Cassava leading the design and delivery of data centres tailored to support AI development in sectors such as healthcare, agriculture and fintech.
Ziaad Suleman, CEO of Cassava Technologies, South Africa and Botswana, says the deal is a major step forward for African enterprise: “Africa has historically contended with receiving second-rate technology.”
He adds that Nvidia’s support will help to change this dynamic and bring high-performance computing to the continent’s AI developers and data-driven industries.
First Nvidia-powered AI data centres in Africa
The partnership launched its first phase in June with the delivery of 3,000 Nvidia GPUs to a Cassava-built facility in South Africa.
A new series of AI factories will be developed over the next three to four years, with a further 12,000 GPUs to be deployed in new facilities across Egypt, Nigeria, Kenya and Morocco.
Eric Omorogieva, analyst at the New Lines Institute, says the introduction of Nvidia hardware into African markets could unlock a step-change in local AI capability.
“Few companies come close to the dominance of Nvidia in the AI sector due to its advances in GPUs,” he says. “Nvidia holds a 93% share in the global GPU market and is the innovator behind most of the advanced chips required to power top systems around the world.
“For African start-ups, access to these chips has been limited due to high costs and lack of availability. By bringing them to the continent, Nvidia will assist in closing the computing gap by facilitating access to hardware throughout key sectors.”
Cassava has also signed a memorandum of understanding with the South African AI Association, providing more than 3,000 AI professionals with direct access to the GPUs. This move is designed to develop human resources alongside technical infrastructure.
Addressing the continent’s compute gap
The initiative is aimed squarely at bridging the continent’s AI infrastructure gap. According to the UN Development Programme, only 5% of African AI practitioners currently have access to the computing power required to carry out advanced work.
Of that 5%, only one-fifth has on-premises access to GPUs, with most relying on limited cloud budgets.
“Addressing this computational gap necessitates specialised infrastructure for AI development,” explains Eric. He adds that the introduction of AI factories will give African organisations better system performance, improved scalability and the ability to turn raw data into actionable outputs.
The data centres will be optimised for AI workloads, offering local developers and institutions high-bandwidth, low-latency access to GPU clusters for model training and inference. This level of capability has been largely unavailable within African borders until now.
Shifting global influence through infrastructure
The Nvidia–Cassava deal arrives at a time of shifting global strategy around African infrastructure.
While China’s Belt & Road and Digital Silk Road programmes have funded state-led initiatives across the continent, the United States is now promoting what the White House calls “commercial diplomacy”.
This strategy, rolled out under the current Trump administration, focuses on private investment and expanded trade to build long-term digital partnerships.
Nvidia’s presence in the African data centre space may also reflect a broader effort by US companies to counter China’s growing influence in AI systems.
While Chinese platforms like DeepSeek R1 and Qwen offer lower-cost models, US companies remain ahead in hardware performance. That performance advantage—especially in GPUs and supercomputing—could be decisive in shaping AI infrastructure globally.
According to Eric, Nvidia’s involvement may help African nations move beyond being passive consumers of AI technology. Access to cutting-edge chips and dedicated infrastructure “will unlock the potential for indigenous African AI systems”, he insists.
