Africa tech

Ghana's $250 million AI bet and Africa's race for compute

Chrisphine Otieno · 19 Oct 2026 · 2 min read

Ghana's $250 million AI bet and Africa's race for compute

On 24 April 2026, President John Dramani Mahama launched Ghana's National Artificial Intelligence Strategy in Accra, according to Ghana's Ministry of Communication. The strategy covers 2025 to 2035 and includes ethical AI, education, workforce readiness, research and ecosystem development. The Ghana News Agency reported plans for a $250 million AI computing centre and $20 million for short- and medium-term implementation. In September, Accra hosted the Pan African AI Summit, where the strategy was on the agenda.

Ghana joins a growing list of African countries with national AI plans, including Kenya, Rwanda, Nigeria, Egypt and others. What stands out is the emphasis on compute.

Why compute matters

AI needs three things: data, talent and computing power. Africa has growing amounts of the first two. The third has been scarce. Most of the GPUs that train and run large models sit in data centres in the United States, Europe, the Middle East and Asia. African users, companies and governments rent that compute, paying in foreign currency and sending their data abroad.

Local compute changes several things:

  • Cost: less foreign exchange spent on cloud services
  • Latency: faster responses for users in the region
  • Data control: sensitive data processed under local law
  • Research: universities and start-ups able to train and test models locally

The sovereignty debate

A Rest of World analysis this year found that many African AI plans still depend on Google, Microsoft, Nvidia and Meta, even as governments talk about sovereignty. That is not necessarily a contradiction; partnerships can build local capacity. But it raises questions about who owns the infrastructure, who sets the terms and what happens when priorities change. Can Africa build AI without big tech?

Kenya's position

Kenya's National AI Strategy 2025–2030 lists AI digital infrastructure as one of its three pillars, alongside data and governance and research and innovation. Where Kenya's AI strategy stands. Kenya's geothermal power has been promoted as an attraction for green data centres. Whether investment arrives at the scale needed is the open question.

Compute is not the whole story

Big computing centres make headlines. The impact of AI on an economy depends just as much on whether ordinary businesses can use AI tools that understand their language and market. A $250 million data centre matters; so does a shopkeeper in Kumasi or Kisumu being able to make a decent advert with AI for the price of a few loaves of bread.

At Tari Studio, we work on that second part: AI images and video for African businesses, tuned for African scenes, people and languages, with costs shown in credits on every quote before anything is made. Africa's AI future needs both the data centres and the everyday tools.