A public-private partnership to develop energy and Artificial Intelligence infrastructure in Africa could be a match made in Heaven.

Artificial Intelligence (AI) has become the holy grail in the global information technology industry overnight, and it appears to be shaping the wealth of nations and the quality of life for people around the world for a very long time to come.

According to UNCTAD, the global AI market is projected to increase 25-fold from $189 billion in 2023 to $4.8 trillion in 2033. On the other hand, AI-related advancements are projected to affect about 15 percent of the global workforce between 2016 and 2030, and 400 million worldwide could be displaced by AI.

The growth of AI is dependent on three important factors: skills, infrastructure, and data. AI infrastructure can be further grouped into two broad categories: computing power (often referred to as “compute”) and digital connectivity (which primarily concerns information and communications technology). Both of these types of AI infrastructure require adequate supplies of affordable water and energy.

The significant growth of the AI industry has led to a substantial and increasing demand for electricity in AI. Although data centers are projected to account for 10% of global electricity demand by 2030, this demand is not evenly distributed worldwide. Thus, data centers will account for 20 percent of electricity demand in advanced economies, compared to 5 percent in emerging and developing economies. As such, advanced economies will face significant challenges in meeting their data center energy needs in the future.

Furthermore, AI-related electricity consumption is projected to increase by up to 50 percent per annum between 2023 and 2030. In the same vein, global data center electricity consumption is projected to double from 536 terawatt-hours (TWh) in 2025 to 1,065 TWh by 2030 due to the increased use of power-intensive AI applications, such as generative AI training and inference.

The United States (US) data centers account for the largest share (45%) of global electricity consumption by data centers, followed by China (25%) and Europe (15%). Furthermore, electricity consumption by datacenters grew by 15% and 12% in China and the US, respectively, between 2015 and 2024. The global increase in data center demand for electricity, coupled with the tendency for data centers to be concentrated in particular locations, means that many jurisdictions have longer wait times (up to 7 years in Northern Virginia in the US) to connect new data centers to the grid. Data centers also consume a lot of water for cooling servers and climate control. Thus, a 100 MW hyperscale data center in the US is estimated to consume approximately 2 million litres of water per day, equivalent to the water use of about 6,500 households. Furthermore, data centers treat the water they use for cooling with chemicals to prevent bacterial growth and corrosion, thus making it impossible to reuse the water in local water supply systems. As a result, many communities are beginning to feel the adverse impacts of hosting data centers on their water supply, and hence many people increasingly oppose the establishment of new data centers in their communities.

Against this background, Big Tech AI companies such as Meta Platforms (Facebook), Alphabet (Google), Microsoft, and OpenAI, which operate hyperscale data centers, are looking for alternative sources of energy, such as nuclear power. While Meta is betting on emerging Small Modular Reactor (SMR) nuclear power technology, Microsoft has entered into a Power Purchase Agreement (PPA) to restart the Three Mile Island nuclear power plant.

Africa has a vast and largely untapped energy resource which, with the right partnerships and investments, can yield huge win-win gains for investors and Africa. Specifically, Africa’s gas, as well as solar and hydroelectric power resources, can produce huge amounts of electricity for power-hungry hyperscale AI data centres.

Africa had 18 trillion cubic metres (tcm) or 8.8 percent of the world’s total discovered natural gas reserves at the end of 2021, and electricity consumption only averages 534.6 kWh per capita compared to a global average of 3,210 kWh per capita. In addition, 598 million (86%) of the 685 million people worldwide without access to electricity live in Africa. Fossil fuels account for almost 72% of the total of over 980 terrawatt hours (TWh) of power generated in Africa, while renewables account for 27%. Hydropower is the largest source of renewable energy in Africa, with 44 GW of installed capacity contributing 19% to the continent’s power generation mix, and up to 90% for Ethiopia and the Democratic Republic of Congo (DRC).

Africa also has huge solar PV and solar thermal potential of over 1 million TWh/yr and 500,000 TWh/yr, respectively. Despite this potential, Africa only generates about 35 TWh and 3.3 TWh of solar PV and solar thermal power, respectively, and accounts for less than 2% of global solar power generation.

Public-Private Partnerships (PPPs) can be effective, win-win vehicles for developing energy infrastructure and data center projects in Africa. Prime candidate countries are Nigeria and Angola, which, between them, have 248 trillion cubic feet of proven gas reserves that can be leveraged to address power shortages. The DRC Grand Inga Dam project, an AU flagship project which will add 6 dams to the existing two dams, will, when completed, generate 44 GW of power per annum. The Grand Inga Dam project needs $80 billion in funding, and, along with many solar power projects in Africa, presents huge opportunities for PPPs to develop Africa’s energy infrastructure to power AI data centers and Africa’s development. Given the Water-Energy-Food nexus in Africa, AI data centers can not only benefit from the abundant water resources of hydropower but also develop groundwater resources, helping to improve water supply for domestic, agricultural, and industrial use.

Despite the global AI divide between industrialized and developing countries, Africa is poised to develop its AI industry and leverage the technology to drive its socio-economic development. Toward this end, the African Union Commission developed a Continental Artificial Intelligence Strategy (CAIS) to harness AI to help achieve Africa’s socio-economic transformation and promote its cultural renaissance, in line with the AU Agenda 2063 (the blueprint for Africa’s transformation), and the Sustainable Development Goals (SDGs). The continental AI strategy is also aimed at minimizing the risks posed by the use of AI in Africa and accelerating the development of AI capacities in Africa.

The CAIS has six Focus Areas, including building “capabilities,” fostering regional and international cooperation, and accelerating AI investment. All of these Focus Areas provide a solid basis for mutually-beneficial partnership and cooperation between Africa and the global AI industry. First, the building capabilities Focus Area can anchor cooperation with Big Tech AI firms in the development of gas-powered, as well as solar and hydro energy infrastructure to power AI data centers in Africa. AI training clusters, which are less sensitive to latency than traditional data centre workloads, are prime candidates for location in data centers based in Africa. If necessary, additional fiber-optic cables can be laid to connect the AI data centers in Africa to data centers in other parts of the world to, among other uses, deploy trained AI models.

Such a partnership between Big Tech companies and Africa will also be in line with the CAIS Focus Areas of fostering international cooperation, and accelerating AI investment. Already, Microsoft announced in 2024 it will invest $1.7 billion and $2.2 billion in Indonesia and Malaysia, respectively, in new cloud and AI infrastructure. In addition, Google announced in 2024 that it planned to invest $2 billion in Malaysia to develop a cloud hub and data center, while Amazon Web Services announced in 2025 its plans to invest $5 billion by 2037 to develop a new hub in Thailand. These companies clearly have the need for building their AI infrastructure and the resources to do so.

Big Tech AI companies should help develop energy and AI data center infrastructure in Africa, especially given the high marginal return on investment in AI in the region. Furthermore, the Big Tech companies are members of the Smart Africa Alliance, which is also aimed at developing AI in Africa and increasing its contribution to the continent’s development. Such investment will be mutually beneficial to the Big Tech companies, who would get access to clean, sustainable and affordable energy to feed new AI data centers, while African countries would benefit from infrastructure that would not only help AI development on the continent but also help solve Africa’s chronic energy shortages.

The business case and moral imperative for the Big Techs to invest in energy infrastructure in Africa are thus very clear. What remains now is the will on both sides to make it happen.