Canada’s Data Centres Are Now Strategic Infrastructure. The Question Is: Who Controls Them, and Are We Moving Fast Enough?

Canada is entering a defining moment in its digital evolution.

Last week, British Columbia quietly signaled a major policy shift. Instead of automatically supplying power to large-scale data centre projects, B.C. Hydro launched a competitive process for 400 MW of electricity. In practical terms, data centres must now justify why they deserve access to the grid.

This is not a temporary constraint. It is an early indicator of a structural reality: compute is becoming one of the most contested resources in advanced economies.

And Canada is not yet acting with the urgency this moment demands.

Compute Is No Longer Just Infrastructure

For decades, electricity powered factories, transportation, and cities. Today, it powers intelligence.

Artificial intelligence models, financial systems, healthcare platforms, defense technologies, identity frameworks, and critical public services all rely on compute capacity. Data centres are no longer passive warehouses for information. They are the operational backbone of national capability.

When a country lacks sufficient domestic compute, it does not merely outsource storage. It outsources leverage.

The implications are economic, geopolitical, and societal.

Who owns the infrastructure increasingly shapes who benefits from the intelligence built on top of it.

Scarcity Is Here And It Changes Everything

A 400 MW allocation would have sounded enormous five years ago. Today, a single hyperscale project can exceed that requirement.

Demand is accelerating faster than grid expansion, environmental approvals, and municipal planning processes can accommodate. Utilities were not designed for the velocity of AI-driven infrastructure growth.

So governments are beginning to triage.

Ontario is prioritizing projects based on economic contribution. British Columbia is evaluating efficiency, heat recovery, and water usage. Not all compute workloads are being treated equally anymore.

This marks a turning point.

Access to power is becoming a policy decision.

And policy decisions inevitably reflect national priorities.

Clean Compute Is Becoming a Competitive Advantage

Canada has long held a structural advantage: abundant renewable energy.

But abundance alone is no longer sufficient. Efficiency is now the differentiator.

Projects that demonstrate:

• advanced cooling

• closed-loop water systems

• heat reuse for communities

• intelligent workload optimization

• lower carbon intensity

are moving to the front of the line.

Clean compute is no longer an environmental talking point. It is rapidly becoming a prerequisite for scale.

The countries that figure out how to deliver sustainable compute at industrial levels will attract the next generation of AI builders.

Canada has the ingredients to lead.

What remains unclear is whether we intend to.

Ownership Matters More Than Many Realize

Here is the uncomfortable question few leaders are asking directly:

If the infrastructure underpinning our digital economy is foreign-owned, how sovereign are we?

Foreign investment is not inherently problematic. Canada benefits from global capital and expertise.

But ownership concentration introduces exposure.

Control over pricing.

Control over expansion timelines.

Control over jurisdictional reach.

Control over operational standards.

When infrastructure becomes strategic, governance matters as much as capacity.

Canada has already learned difficult lessons about infrastructure ownership. Nearly two decades ago, it became clear how exposed we were when significant portions of our hydro-electric assets fell under foreign control.

Compute deserves the same level of scrutiny not after concentration occurs, but while the landscape is still being shaped.

Data residency alone is not sovereignty.

Operational authority is.

The Greater Threat Is Structural, Not Just External

The greatest threat to Canada is not only malicious actors. It is hesitation, particularly hesitation to foster a competitive, diversified infrastructure landscape.

If we are not deliberate, we risk replicating the concentration dynamics that have long defined Canada’s telecommunications sector, where limited competition slowed innovation and constrained national leverage.

This moment calls for urgency, but also intentional market design.

Permitting cycles stretch for years.

Grid upgrades lag demand forecasts.

Regulatory frameworks struggle to keep pace with technological acceleration.

Meanwhile, other nations are treating compute as a national priority.

Delay has consequences.

If builders cannot scale here, they will scale elsewhere.

If innovators cannot access domestic compute, they will train models abroad.

If critical workloads migrate, so does influence.

By the time shortages become obvious, strategic dependence is already locked in.

Infrastructure decisions echo for decades.

Moving quickly does not mean abandoning standards. It means ensuring that sovereign, Canadian-owned builders, including emerging small and mid-sized operators, can realistically enter the market.

Access to capital, power allocation frameworks, and regulatory pathways must reflect this priority. Otherwise, the future of Canadian compute risks being decided by a narrow set of incumbents before the ecosystem has a chance to mature.

This Is Bigger Than Technology

Data centres sit at the intersection of several national objectives:

• economic resilience

• energy strategy

• climate commitments

• defense readiness

• privacy protection

• innovation competitiveness

Treating them as routine commercial developments underestimates their impact.

Canada needs a coordinated compute strategy that aligns federal, provincial, and utility leadership.

Not in five years. Now.

What Leadership Should Look Like

If Canada intends to remain digitally self-determined, several shifts must happen quickly.

1. Recognize compute as strategic infrastructure. Place it in the same planning horizon as transportation and energy corridors.

2. Accelerate approvals without lowering standards. Speed is now part of competitiveness.

3. Incentivize Canadian ownership and partnerships. Ensure domestic participation in the assets that will power our economy.

4. Tie power allocation to long-term national benefit. Prioritize projects that create ecosystems, not just facilities.

5. Build for efficiency from day one. Clean compute should be engineered, not retrofitted.

6. Align energy expansion with AI demand forecasts. Planning based on yesterday’s consumption curves will fail.

7. Design for competitive depth. Encourage a landscape where multiple Canadian operators can scale, reducing systemic dependence on a small number of providers.

Opportunity Is Still Within Reach

Canada is not behind, but the window to shape ownership, competition, and sovereignty is narrower than many assume.

We possess renewable energy, political stability, strong research institutions, and growing AI talent. Few countries have this combination.

What we require now is conviction.

The shift already underway in British Columbia is more than a provincial procurement exercise. It is a signal that compute is entering a governed era.

The question facing Canadian leaders is straightforward:

Will we shape this infrastructure intentionally, or inherit whatever the market happens to build?

Because in the age of AI, the countries that control compute will help define the future.

And sovereignty, once surrendered, is rarely reclaimed.

The window to act is still open.

But it is narrowing.