Adam McCabe

← Back to home

Securing Canada's Future: The Case for Sovereign AI

February 22, 2025

I've been thinking a lot about Canada's place in the world lately. Not just as a dad raising my son, Emery, who's five, or as someone working in AI, but as a Canadian watching the landscape shift in ways that feel increasingly dangerous. If the last few years have taught us anything, it's that being a nice, resource-rich, and strategically located country isn't enough. If you don't have deterrents—military, economic, technological—you're at risk of becoming someone else's asset.

Trump's 2025 comments about Canada's resources, annexation, and strategic position weren't just another bit of hyperbole. When the most powerful country in the world starts casually discussing its northern neighbor as something to take advantage of, we should pay attention. This isn't just rhetoric. This is a sign of a global realignment where old assurances—like NATO, U.S. defense commitments, and economic partnerships—are becoming increasingly transactional. And Canada, in its current state, isn't prepared to be in that game.

The Ukraine Lesson: Security Promises Are Fragile

I keep coming back to Ukraine. In the 90s, they had nuclear weapons, but they gave them up for security assurances from Russia, the U.S., and the UK. Less than 30 years later, those assurances were worthless. Russia invaded, and Ukraine was left scrambling for military aid, weapons, and technology. They didn't have deterrence, and they paid the price.

Canada's situation is different, of course. We have NATO. We have alliances. But what happens when our biggest ally shifts its priorities? If we're relying on the U.S. for our strategic security while the same U.S. government talks about taking our resources, something isn't adding up. And if the last decade has taught us anything, it's that we need to be thinking about resilience, deterrence, and sovereignty in ways we never had to before.

ASI Is the Next Great Power Move, and We're Behind

Artificial Superintelligence (ASI) isn't just an abstract concept anymore. We're on a trajectory where the first country to reach ASI will have a strategic advantage on the level of nuclear weapons in the 1940s. It will reshape economic, military, and geopolitical power. And right now, the two most likely candidates to get there first are the U.S. and China.

Let's be blunt: neither country is going to share their most advanced AI models with us. The idea that we can coast on our proximity to the U.S. and just benefit from their AI breakthroughs is naive. The most powerful AI systems won't be open-source. They won't be freely accessible. They will be classified, controlled, and used for national advantage. If Canada isn't building its own, we're going to find ourselves locked out of the most important technological shift of our lifetime.

We've been lucky. Canada has been punching above its weight in AI research for years. The University of Toronto, Mila in Montreal, and Amii in Edmonton have produced some of the most foundational AI research in the world. We have the talent. We have the expertise. But we don't have the infrastructure.

Most large-scale AI model training doesn't happen in Canada. Our researchers rely on Google, NVIDIA, and other U.S.-based cloud providers for compute. Our own AI companies—even the successful ones—often have to use American infrastructure to train their models. Cohere, one of our biggest AI startups, is only now getting support to build a Canadian-based AI data center. But that's still one company. Meanwhile, the U.S. and China are pouring billions into AI supercomputing and sovereign AI stacks. We're just starting to think about it.

The Economic Case for AI Sovereignty

Even beyond defense, Canada's economic future depends on our ability to develop, train, and deploy AI models domestically. Right now, our economy is facing serious headwinds. Governor of the Bank of Canada Tiff Macklem recently warned that, under sustained tariffs, Canada could see a cumulative 8% economic contraction over 2-3 years. If tariffs come down hard, if global trade tightens, Canada is going to need new engines of economic resilience. The AI pipeline should be one of them.

But we can't build an AI-powered economy without investment. The government has started to move in the right direction—a $2 billion plan for sovereign AI computing infrastructure, funding for data centers—but it's not nearly enough. We need a national AI strategy that isn't just about research grants. We need policy that makes it easier for companies to build, train, and deploy AI models in Canada, at scale. That means investment in compute infrastructure, partnerships between industry and government, and a recognition that AI isn't just an industry—it's a national security issue.

What Canada Needs to Do—Now

We don't have the luxury of time. The world isn't going to wait for Canada to figure this out. If we want to be a serious player in AI and national defense, we need to make moves now:

Sovereignty Means Innovation, and Innovation Means Security

This isn't just about AI. It's about making sure Canada isn't left behind in the next great power shift. If we don't act, we're going to find ourselves in a world where the U.S. and China control the future of AI, and we're left hoping they let us use it. Hope isn't a strategy.

Canada has the talent, the resources, and the economic strength to lead in AI. But we need to stop thinking of AI as just another tech industry and start treating it as what it really is: the next frontier of national security and economic power.

We can either build it ourselves—or wait for someone else to decide how we get to use it. I know which future I'd rather bet on.