It is widely recognized that the Fourth Industrial Revolution will be driven by technological advancements in Artificial Intelligence, Cloud Computing, Internet of Things and Robotics. Countries and tech companies are investing heavily in emerging technologies as these will provide a decisive advantage in multiple domains. However, AI will require vast amounts of energy to power data centers and run machine-learning algorithms. This technological shift demands not only a reliable energy supply but also one that is scalable and sustainable. In this context, nuclear energy has emerged as a viable alternative to power the technologies of the future.
Data centers are referred to as the “brains” of the Internet. Global giants like Microsoft, Amazon, Google, Apple and Meta are racing to build data centers that provide the necessary infrastructure to train and deploy complex machine-learning models and algorithms. These tech companies are particularly focusing on data centers that are well-suited for data-intensive tasks such as generating AI responses.
According to the US Department of Energy, the energy requirement for a large data center is 100 MWe, which is equivalent to the energy consumed by approximately 80,000 households in the United States. In a data center, the process of cooling the servers and Central Processing Units (CPUs) consumes the largest share of electricity. McKinsey and Company, a global consulting firm, estimates that the process of cooling accounts for nearly 40 percent of the total energy consumed by AI data centers. In addition, computing accounts for 40 percent of the electricity while 20 percent is used for IT-related equipment in a data center.
A report titled “Electricity 2024-Analysis and Forecast to 2026” by International Energy Agency (IEA) highlights that global energy demand from data centers will double by 2026. In 2022, data centers consumed an estimated 460 terawatt-hours (TWh) globally and this figure could increase to more than 1000 TWh in 2026. A measure of how much electricity is consumed by AI data centers can be gauged by the fact that this increased demand is roughly equivalent to the annual electricity consumption of Japan. Another indication of AI’s energy needs is the fact that OpenAI’s ChatGPT requires around 10 times as much electricity as compared to a Google search.
There are currently more than 8,000 data centers globally, with about 33% located in the US, 16% in Europe and around 10% in China. The electricity consumption from data centers in the US is expected to increase from around 200 TWh in 2022 which accounts for 4 percent of total electricity demand in the country, to almost 260 TWh in 2026, accounting for 6% of the total electricity demand. The Electric Power Research Institute (EPRI) projects that data centers could consume up to 9% of electricity in the US by 2030.
China’s rise as a global technology leader is largely supported by its data center ecosystem, with tech companies looking to scale up their data centers to ensure the reliability and stability of data services. According to a report from China’s State Grid Energy Research Institute, electricity consumption by data centers in China will exceed 400 billion KWh, accounting for 3.7 percent of the country’s total electricity consumption, by 2030. It is also estimated that electricity consumption by data centers in the European Union will increase from 100 TWh in 2022 to approximately 150 TWh by 2026. The IEA report highlights that by 2026, almost one-third of electricity demand in Ireland is expected to come from data centers.
As the demand for electricity increases, traditional renewables like solar and wind will be unable to provide the uninterrupted power these tech companies require. In addition, companies cannot rely on energy from fossil fuels because of their large carbon footprint and adverse impact on climate. Therefore, tech companies are investing in Small Modular Reactors (SMRs) which are next-generation nuclear reactors with a capacity of up to 300 MWe. SMRs are feasible due to their scalability, lower upfront costs and enhanced safety features as compared to conventional large-scale reactors.
Tech leaders such as Bill Gates, Jeff Bezos, Elon Musk, Larry Ellison and Mark Zuckerberg have emphasized that nuclear energy is the only viable solution to keep pace with the growing energy needs of AI data centers. The International Atomic Energy Agency (IAEA) in a report released in September 2024 titled Electricity and Nuclear Power Estimates for the Period Up to 2050 projected that global nuclear capacity will increase 2.5 times by 2050, with SMRs playing a key role in this growth.
To satisfy the growing energy needs of AI data centers, Big Tech companies have announced that they will invest in reviving existing nuclear power plants and in developing next-generation nuclear reactors, focusing on SMRs. Amazon is building SMRs and has signed a $500 million deal with the X-Energy Reactor Company. Amazon and X-Energy are also collaborating to bring more than 5000 MWe of new power projects online across the US by 2039, representing the largest commercial deployment target of SMRs.
Similarly, Google has set a ‘moonshot goal’ of running its data centers entirely on carbon-free energy by 2030. Google has also partnered with Kairos Power to bring 500 MWe online by 2035, with the first SMR expected to be operational by 2030. Microsoft and Constellation Energy have also signed a $1.6 billion deal to restore the Three Mile Island nuclear plant which will help secure carbon-free energy for the next 20 years. In addition, Oracle is designing a data center that will require more than a thousand MWe of electricity, which will be powered by three SMRs. These announcements by Microsoft, Google, Amazon and Oracle demonstrate the willingness of tech giants to finance the development and deployment of SMRs for their own energy requirements and climate goals.
The progress in building data centers is still in its infancy in Pakistan, as these centers require a well-developed digital infrastructure and extensive access to fiber optic cable networks. According to Data Center Map, Pakistan currently hosts 22 data centers which are based in Karachi, Lahore and Islamabad. Companies and government departments including NADRA, PTCL, Jazz, Ufone and the National Information Technology Board (NITB) are operating these data centers. One positive development is that in April 2024 a software company from the Czech Republic “IceWarp” announced an investment of USD 1 million to launch a data center in Karachi. To enhance its data center ecosystem, Pakistan needs to attract further foreign investment by offering incentives like affordable land and favorable policies, thereby paving the way for digital transformation and economic growth.
As the number of data centers grows in Pakistan, the energy requirement to power these centers will also increase. A stable and reliable power supply is crucial for the efficient operation of these data centers, making SMRs a viable solution. Under its Nuclear Energy Vision 2050, Pakistan aims to produce 42,000 MWe by 2050, which is expected to meet one-fourth of the country’s energy needs. This target can be achieved by investing in SMRs which are a feasible option for Pakistan given their low cost and reduced construction time as compared to large conventional reactors. SMRs have the added benefit of operating independently from large power grids, making them a suitable option for providing electricity to rural and far-flung regions in Pakistan.
SMRs can also help Pakistan generate revenue through carbon credits due to their low-carbon footprint. In November 2024, the Ministry of Climate Change and Environmental Coordination launched the ‘Pakistan Policy Guidelines for Trading in Carbon Market’ during COP29 in Baku. The policy aims to decarbonize Pakistan’s economy and generate revenue by selling carbon credits. Countries can earn carbon credits by engaging in activities that reduce emissions, such as investing in clean energy including nuclear and renewable energy and implementing energy efficient measures.
In order to catch up with rest of the world and excel, Pakistan needs to consider establishing Data Center Outsourcing (DCO) based on the global practice of colocation. The concept of colocation involves the outsourcing of data centers where a company rents physical space from a third-party provider for the storage of their data. According to Straits Research, the global data center colocation market is estimated to grow to USD 159.8 billion by 2030.
Given the potential revenue that can be earned from DCO, many countries across the world are offering incentives to attract foreign investment. For example, Central Asian countries like Kazakhstan and Uzbekistan are inviting investment through providing colocation services. Similarly, Pakistan can benefit from DCO by establishing dedicated tech cities, powered by SMRs. Investment for such tech cities could be invited from Pakistan’s investment partners, in particular Saudi Arabia, the UAE and China.
Be the first to comment