The EU - Canada push for greater AI sovereignty

Brussels is sharpening its security posture for AI. Ottawa is scaling its national Artificial Intelligence (AI) strategy. Where can Canadian - Swedish partnerships reinforce both?  

The answer matters more than ever. AI sovereignty is now a core pillar of economic resilience, security, digital infrastructure, and geopolitical alignment; and it’s a priority for governments everywhere. Canada and the EU are among them, investing in sovereign compute, tightening security frameworks, and building partnerships that reduce reliance on foreign cloud and AI platforms.

A strong Canada-Sweden partnership has a role to play in this agenda. Government-level commitments only hold up if businesses and sectors expected to run on sovereign systems are engaged in building them. Without that, sovereignty stays a policy intention rather than something transferable in practice.  

The partnership between Canada’s Mila and Sweden’s Rise could guide us on how it’s done. 

EU’s Action Plan for building secure and sovereign AI 

Signed in December 2025, the EU-Canada MoU on AI commits both sides to align cooperation with the EU's Apply AI Strategy and Canada's national framework, and to keep exploring further work with like-minded partners.  

The European Commission added to that momentum by publishing in July 2026 its Action Plan on Cybersecurity and Artificial Intelligence. It is a coordinated framework to help member states, businesses, and public authorities manage the security risks that come with advanced AI models. To this end, the Commission states that the EU must continue investing in its own sovereign advanced AI capabilities, drawing on the infrastructure of its AI Factories. 

Canada’s ‘AI for all’ labelled a sovereignty strategy 

On June 4, Prime Minister Mark Carney launched "AI for All," Canada's new national AI strategy, built around trust, opportunity, and sovereignty. The plan pairs a "build-partner-buy" approach to infrastructure with a goal to close Canada's AI adoption gap, while working to reduce reliance on foreign cloud and compute providers along the way. 

Canada is also reworking its regulatory foundations. The federal government has begun developing a new AI governance framework, one that reflects generative AI realities and supports Canada’s sovereignty goals around trusted, secure, domestic AI capability.  

Sovereignty lies at the heart of Canada’s AI for All strategy: “Canadians want safe, reliable, and sovereign AI. They want the best tools to build a prosperous future guided by our values. That is our plan.”

Can sovereignty be shared? 

A framework is only as strong as what gets built underneath it. That’s where Sweden-Canada partnerships can play a role.  The partnership between Montreal's Mila, co-founded by Université de Montréal and McGill University, and RISE, Sweden's national research institute, offers a live example.

Their April MoU, backed by a joint steering committee, aims to apply AI to forestry, mining, and energy systems, with human-centered trust built in as a design requirement. This is a practical model for how sovereignty is an integrated priority, co-developed without compromising national policy. 

Sovereign AI capability doesn't come from infrastructure alone; it depends on industries trusting and adopting the systems built for them. Trust is what turns a domestic AI capability into one that companies and institutions rely on instead of defaulting back to foreign platforms. In that sense, what Mila and RISE are building in forestry, mining, and energy is sovereignty work, even if neither organization frames it that way.   

At the Canadian Chamber of Commerce in Sweden, we’re tracking these kinds of partnerships closely. AI and technological sovereignty are topics covered in Pillar 4 of the Canada-Sweden Strategic Partnership.  

Home-grown AI. The end game for both EU and Canadian AI strategies.