Canada's Dependence on U.S. AI Providers Raises Sovereignty Concerns

A recent study from the University of Toronto has brought to light a significant discrepancy between the Government of Canada’s stated intention of fostering sovereign artificial intelligence capabilities and its actual procurement practices. The report indicates that the Canadian federal government remains heavily reliant on U.S. technology providers, particularly Microsoft, for its AI system requirements. This reliance raises critical questions about the diversity of procurement sources, the accountability of technology choices, and the potential for long-term dependencies that could affect the government’s operational integrity and policy goals.

Published in August 2026, the research by Dipto Das, a postdoctoral fellow at the University of Toronto, scrutinizes the Federal AI Register established by the government in November 2025, designed to promote transparency in AI usage within public services. According to the findings, of the AI systems noted in this register, a striking 46% were sourced from Microsoft. This disproportionate concentration of procurement raises alarm bells about the health of the local tech ecosystem and undermines the aspirational goals of developing a robust internal capacity for AI solutions. The study highlighted that 38.1% of the systems in use came from external vendors while 43.3% were developed in-house, leaving only 6.1% for open-source alternatives.

The implications of this vendor concentration are profound. As many government procurement professionals strategize to align their sourcing decisions with national sovereignty objectives, they must also navigate the economic realities that often favor established players like Microsoft. The University of Toronto study suggests that the current framework may inadvertently prioritize successful procurement processes focused on efficiency over the diversity of suppliers and technological accountability. Das remarks, "As we are saying those things, are we really putting those things into practice? In most cases, we found that there are a lot of discrepancies." This statement emphasizes the growing gap between the government's public affirmations of sovereignty and the underlying procurement practices that compromise true independence.

Moreover, with 86% of the recognized systems being deployed for efficiency and workload reduction, it’s evident that there’s a prioritization of operational effectiveness over broader public interest outcomes. The study critiques the Federal AI Register for lacking accountability and meaningful oversight — two fundamental aspects that could enhance public trust in technology’s role within federal services. The AI and Data Act, which aimed to establish binding regulations on private-sector AI, remains stalled, leaving a vacuum where comprehensive governance ought to be. The situation poses broader questions: Can strategic sovereignty be achieved amid heavy reliance on foreign technology providers?

As the Canadian government presses forward with its digitization strategies amid budget cuts and job reductions, the concerns grow over how dependency on foreign cloud services will influence its capacity to maintain operational autonomy. Contracting experts and vendors within the government contracting sphere would do well to monitor these developments, considering potential shifts in public procurement policies that could open doors for domestic innovations.

This scenario is particularly cautionary, not just for Canada, but for other governments experiencing similar reliance on dominant U.S. technology firms. Countries like the United Kingdom and Australia struggle with analogous dependencies, confronting the challenges presented by reliance on a limited number of global technology providers for public-sector digitization and AI services. As governments worldwide navigate these turbulent waters, cultivating a diverse supplier base in critical sectors like AI will be essential for securing both economic stability and strategic independence.