ARTICLE
3 November 2025

Data Centres And Artificial Intelligence: A Growing Opportunity For Canadian Nuclear Power?

F
Fasken

Contributor

Fasken is a leading international law firm with more than 700 lawyers and 10 offices on four continents. Clients rely on us for practical, innovative and cost-effective legal services. We solve the most complex business and litigation challenges, providing exceptional value and putting clients at the centre of all we do. For additional information, please visit the Firm’s website at fasken.com.
The International Energy Agency (IEA) released a special report about "Energy and AI" in 2025. The report focuses on the consequences of the emergence of artificial intelligence (AI)...
Canada Energy and Natural Resources
Dierk Ullrich’s articles from Fasken are most popular:
  • within Energy and Natural Resources topic(s)
  • with Inhouse Counsel
  • with readers working within the Healthcare, Technology and Metals & Mining industries

The International Energy Agency (IEA) released a special report about "Energy and AI" in 2025. The report focuses on the consequences of the emergence of artificial intelligence (AI) in the global energy sector, specifically electricity. Among the mix of energy sources needed to deliver the uninterrupted power supply that data centres need to support AI, there are established clean energy technologies such as wind, solar, and hydropower. Canada can also supply nuclear power, including emerging technologies like small modular reactors (SMRs).

The report notes that on a global basis "data centre electricity consumption has grown by around 12% per year since 2017, more than four times faster than the rate of total electricity consumption." The IEA estimates that, by 2035, more than 900 terawatt hours (TWh) in new electricity generation will be needed to support the growing power demand of data centres. Half of the global growth is expected to be met by renewable energy. Natural gas is expected to meet one quarter of the additional demand, with nuclear power estimated to account for the remaining quarter. With the first grid scale SMRs in Western countries expected to go online in 2030 (see our bulletin: Another Sign of Canadian Leadership in Nuclear Energy: North America's First SMR Has Received Federal and Provincial Approval) a combination of conventional and emerging nuclear energy sources may become available just in time, in particular in the Canadian market, to meet the growing power needs of AI and data centres.

This is consistent with demand growth expectations in Canada. For example, the Ontario Independent Electricity System Operator has forecasted that electricity demand in the province will grow by approximately 75% over the next 25 years, rising from 151 TWh in 2025 to 263 TWh in 2050. This was a significant upward revision from the previous year's projection of 60% growth. A key driver of this expected growth is new and larger data centres in the province to support AI and cloud computing services. Alberta, British Columbia, Québec, and other jurisdictions in Canada are anticipating similar demand growth.

Procuring Nuclear Power for Data Centres and AI

The IEA report outlines some of the procurement strategies employed by technology companies seeking to meet the vast electricity supply needed to meet data centre demand. In touching on nuclear, the IEA notes:

"As part of these strategies, technology companies are also supporting the development and commercialization of innovative low-emission base load technologies, such as small modular reactors (SMRs).... To date, plans to build up to 25 GW of SMR capacity associated with supplying the data centre sector have been announced worldwide[...]. The first projects are expected to start to materialize only towards the end of this decade."

A key challenge for nuclear power in meeting the electricity demand of data centres and AI is the lengthy permitting and construction process in comparison to other sources. For some conventional sources such as wind, solar and natural gas generation, permitting and construction can be completed in as little as five years. By comparison, for nuclear power, the IEA report states:

"Other dispatchable technologies, such as large-scale nuclear reactors or hydropower plants, typically require closer to a decade or more to complete. Once SMRs... become commercial, they may also offer medium-length development times of approximately three to five years."

Developments in Canada

Canada has embraced SMRs as part of its nuclear strategy. In April 2025, the Darlington New Nuclear Project became the first SMR under construction anywhere in the world (outside of China or Russia). It is the first of four grid-scale SMRs that Ontario Power Generation plans to construct at the Darlington site. The goal is to complete the first of these and provide power to the grid by the end of 2030. (For additional details see our bulletin: Another Sign of Canadian Leadership in Nuclear Energy: North America's First SMR Has Received Federal and Provincial Approval).

More recently, in September 2025, the federal government named the Darlington New Nuclear Project as one of the first five projects of national interest under the recently enacted Building Canada Act. This new federal legislation aims to streamline the regulatory approval process for projects of national interest. Another goal of the new legislation is to facilitate more efficient private sector financing of these projects by declaring them of national interest, therefore securing the feasibility of the project.

If Canada can find efficiencies in developing and financing SMRs and other nuclear energy projects, this will play a significant role in addressing the forecasted growth in electricity demand for data centres and AI, not only in Ontario but throughout Canada.

Developments in Other Parts of the World

The IEA report provides helpful guidance on how nuclear power could help meet the growing electricity demand from data centres and AI in other jurisdictions beyond Canada. The IEA writes that nuclear power will play a significant role in meeting data centre electricity demand.

Since release of the IEA report, we have seen additional developments supporting the aggressive advancement of nuclear power projects which include the following:

  • In the United States, many of the tech giants and large data centre operators have recently announced a range of projects involving planned collaborations with SMR developers to site nuclear power projects near, and supply electricity to, planned data centres.
  • In Europe, renewable and nuclear are expected to supply most of the additional electricity required.
  • In China, data centres are located mostly in the east of the country and their electric supply is dominated by coal. After 2030, the introduction of SMRs is expected to significantly boost the nuclear share of the data centre electricity mix.
  • According to the IEA, Japan and South Korea together currently account for approximately 5% of global data centre demand. Renewable and nuclear energy are set to provide approximately 60% of the electricity consumed by data centres in 2035, up from 35% now.
  • The rest of the world accounts for approximately 10% of total data centres electricity generation today, primarily in Southeast Asia and India.

AI and Energy Innovation

Much of the excitement around AI, and the rush to develop energy infrastructure and data centres to support it, lies in the anticipated efficiencies and innovations that AI may facilitate. As part of its report, the IEA addresses how the growth of AI may accelerate the innovation of nuclear energy technologies. It notes that the world is entering a new "Age of Electricity" and that interest in nuclear power has grown to a 50-year high. It adds:

"The technology sector is making important investments and commitments to nuclear power [...]. New technologies such as small modular reactors (SMRs) remain to be demonstrated at scale but hold promise for the fast-growing industry because they have lower upfront investment than conventional plants, which could be more attractive to private investors (most SMRs under development are expected to cost less than USD 2 billion, compared to greater than USD 10 billion for conventional nuclear in some markets). However, high levelised costs and regulatory hurdles remain significant barriers to deployment."

Recent Related Canadian AI Policy Developments

Recently, the Minister of AI and Digital Innovation launched Canada's AI Strategy Task Force, with a mandate to renew the Canadian National AI Strategy.

Earlier in 2025, the Canadian federal government released the Canadian Sovereign AI Compute Strategy, announcing that the sum of $2 billion over five years has been allocated to launch new AI initiatives. One pillar of the strategy is to mobilize private sector investment.

In order to achieve this, the federal government has put together the "AI Compute Challenge", a government program that will seek out, among other initiatives, project proposals to build out or expand commercial AI-specific data centres in Canada.

Conclusion

The growing global demand for new electricity generation, driven in large measure by the apparently insatiable energy needs of data centres and AI, is creating challenges and opportunities across the energy sector. Nuclear power is expected to play an important role in meeting this demand. Canada has made early commitments to the future growth of nuclear power to address our domestic energy needs, especially in the area of AI and data centres. As an investment location, Canada also offers competitive advantages with respect to land availability, climate, and other essential resources, such as access to water. With ambition, vision, and careful execution Canada could reach the end of the decade with not only its first operating grid-scale SMR, but with many more nuclear power projects in development to address our growing need for electricity.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.

Mondaq uses cookies on this website. By using our website you agree to our use of cookies as set out in our Privacy Policy.

Learn More