Procurement Is the Lever
The fastest path to sovereign AI in Canada is procurement reform. When federal and provincial governments spend billions on cloud services and AI infrastructure, that spending should mandate Canadian ownership — not just Canadian data residency, but Canadian legal and operational control of the infrastructure itself. Alberta has taken the lead, issuing procurement requirements that explicitly define sovereign compute as environments free from foreign legal authority. Every province should follow.
Trade Agreements and Data Sovereignty
International trade agreements increasingly include digital trade chapters that limit governments' ability to require local data storage. Canada must negotiate carve-outs that protect sovereign AI infrastructure for sensitive government data — healthcare, justice, corrections, defence, Indigenous governance. Data sovereignty is a national security issue, not a trade barrier. The European Union has demonstrated that strong data protection regimes are compatible with international trade.
Investment in Canadian Capacity
Sovereign AI requires three investments: physical compute infrastructure, open-source software ecosystems, and human capital. Canada has world-class AI research talent, but that talent works overwhelmingly on foreign-owned platforms. Redirecting even a fraction of government AI spending toward domestically owned infrastructure creates a virtuous cycle — Canadian talent building Canadian capacity for Canadian needs. Jerald Sibbeston and Yamoria are building from the ground up in the Northwest Territories, proving the model works.