Canada wants sovereign AI data centres. What does that actually mean?

Canada is advancing plans for sovereign AI data centres to decrease dependence on U.S. cloud providers like Amazon and Microsoft, which currently host much of its digital infrastructure. The federal government is reviewing over 160 proposals and has allocated $925.6 million over five years to support large-scale public AI infrastructure, with a focus on keeping data subject to Canadian laws. Telus became the first successful applicant under the program, though negotiations and funding commitments remain ongoing. Critics argue that true sovereignty is difficult to achieve, as data centres may still rely on foreign hardware like Nvidia’s GPUs and global networks that transcend national borders. A recent lawsuit against the U.S. Department of Homeland Security highlighted concerns over foreign governments accessing Canadian data, reinforcing Ottawa’s push for domestic control. Minister of Artificial Intelligence and Digital Innovation Evan Solomon emphasized that new data centres will have sovereignty requirements but clarified that operators won’t be restricted to serving only Canadian governments or industries. Bell Canada, another major player, is investing $1.7 billion in a new 300-megawatt data centre near Regina, signaling growing momentum in Canada’s AI infrastructure sector. Industry leaders, including Telus CEO Mirko Bibic, expressed optimism about the government’s push to develop AI infrastructure domestically. Bibic stated that Canada could build ‘something that’s built and made in Canada, for Canadians, in a sovereign way,’ though challenges remain in ensuring full control over data flows and foreign dependencies.