Is Canada’s AI Strategy Moving Fast Enough to Compete Globally and Protect Nationally?

Canada, despite becoming the first country to announce an AI strategy in 2017 and investing in research and talent, which initially led the tone in the global race, now finds itself lagging in adoption and another even more critical piece of the puzzle—computing power. The systems that train and deploy large-scale AI models require advanced chips, massive data centers, and long-term investment.

Currently, the two largest AI players, the U.S. and China, as well as some smaller countries with aggressive industrial strategies, are outpacing Canada on that front.

In July, President Donald Trump’s administration officially announced “America’s AI Action Plan,” prioritizing deregulation in a bid to strengthen U.S. leadership. Critics argue this favours industry over public safety, a sharp contrast with Europe’s strategy of building AI guardrails around safety, fairness, and accountability.

According to a recent poll, Canadians seem more in line with the European approach, finding 85% of Canadians support regulating AI, with 57% strongly in favour, though views remain split on whether AI is good or harmful for society.

Canada, however, has yet to deliver. While it has the talent, its infrastructure is underdeveloped, its regulations are stalled, and the gap raises a pressing question: Is Canada moving fast enough to match both the ambitions of its allies and the concerns of Canadians?

What is Canada’s AI Strategy?

In December 2024, facing pressure over its lack of AI infrastructure, the federal government unveiled the Canadian Sovereign AI Compute Strategy, a plan to allocate CAD 2 billion toward enhancing the country’s AI capabilities.

In May, Canada appointed its first Minister of AI and Digital Innovation, Mr. Evan Solomon, a rookie MP and former journalist representing Toronto Centre. Many view this as a positive step, marking us as the third country to appoint a government official to a dedicated role that oversees AI initiatives, following France in 2024 and the U.A.E. in 2017.

Adegboyega Ojo, Canada research chair in Governance and Artificial Intelligence at Carleton University’s School of Public Policy and Administration, was quoted as saying, “The creation of the Ministry of AI and Digital Innovation signals the new government’s intention to prioritize AI development as part of its broader ambition to build an economy of the future.”

The sentiment is shared by some software engineering leaders with teams spread across global offices, seeing the varying approaches each country has in developing and regulating the technology, including Ranjit Tinaiker of Ness Digital Engineering, with offices based in Ontario and abroad, who sees the creation of the ministry as a sign that Canada intends to play a larger role in shaping how AI is used and governed.

“2025 is the year that leaders need to transition from treating AI as the new shiny object… to taking concrete steps to adapt to the incredible impact that this technology can deliver,” adds Tinaiker.

Newly elected Prime Minister Mark Carney named AI a national priority, pointing to its potential to improve government services, generate new jobs for Canadians, and seek global partnerships to build the technology. It was our first real glimpse into the new government's plans on the topic.

In the minister's first speech, he stressed that the government’s priority in terms of regulation will be to protect data and privacy, not over-regulate innovation.

In an effort to offer transparency on how AI is being used at the federal level, the Liberal government announced a public registry listing the AI projects being used to scale efficiency in various public service departments.

Guardrails are missing on both sides of the border.

Some experts argue that the Liberal AI plan focuses narrowly on AI’s promise, without first investing in creating regulations to mitigate its potential impacts.

“Much of the Liberal plan seems to involve taking risks. There’s a shortsightedness on this rapidly advancing technology that requires significant guardrails,” says Jake Pyre, PhD Candidate at Concordia University, whose research focuses on the futures that are sold to us by tech companies.

“The government seems to view AI as a solutions machine, buying into the hype around it without taking the time to understand it.”

This lack of focus on guardrails is also reflected in the slow progress of key legislation. The Artificial Intelligence and Data Act, the Online Harms Act, and the Act Respecting Cyber Security were all introduced to shape how AI and digital systems operate in Canada, but none are in force today.

The first two were dropped when Parliament was prorogued earlier this year, while the cybersecurity bill had to be reintroduced as Bill C-8 and remains at second reading.

In June, Minister Solomon said regulation isn’t about finding “a saddle to throw on the bucking bronco called AI innovation. That’s hard. But it is to make sure that the horse doesn’t kick people in the face. And we need to protect people’s data and their privacy.”

The U.S. faces its own version of this gap. While the federal government is pouring resources into AI infrastructure and chip manufacturing, it has yet to introduce binding rules on its use in the interest of mitigating its impacts. Trump’s AI agenda, in particular, has drawn backlash from more than 90 technology, labour, environmental, and civil society groups, who accuse the administration of prioritizing Silicon Valley’s corporate interests over public safety and oversight.

“We need rules and accountability— not a Silicon Valley free-for-all,” said J.B. Branch of the U.S.-based advocacy group Public Citizen.

Building AI at home: The push for sovereign control

In today’s world, the value of national ownership is viewed more in the interests of security and economic nationalism, especially as geopolitical tensions rise.

As part of Canada’s objectives, the plan looks to build supercomputing infrastructure and hardware through a mixture of private and public investment and partnerships within its borders, by giving domestic companies the tools and capital to train, test, and deploy competitive AI models so that the country can rely less on foreign-owned cloud giants, like Amazon and Microsoft.

As of now, many Canadian companies implementing AI are reliant on subscribing to cloud-based AI Computing vendors outside of the country, presenting challenges related to growth, data sovereignty, and talent retention.

"If we have a data centre that's, let's say, storing our health-care data, but it's located inside the United States, things like the Patriot Act and the Cloud Act allow the American government to access it," says Benjamin Bergen, president of the Council of Canadian Innovators.

In May, Telus, on of Canada’s biggest telecom providers, announced it would be investing more than $70 billion in Canada over a five-year period to expand its network, including launching two new AI data centres. Cohere, one of Canada’s leading AI firms, secured $240 million to deploy a data centre under the nation's compute strategy, to be built in partnership with U.S.-based GPU/cloud provider company CoreWeave.

In August, Prime Minister Mark Carney announced the creation of the Major Project Office (MPO), headquartered in Calgary with plans to expand to other cities. The new office will coordinate financing and streamline approvals to accelerate the development of the “nation-building project”.

“This would build compute capacity and data centres that we need to underpin Canada’s competitiveness, to protect our security, and to boost our independence and sovereignty,” Carney said. “This will give Canada independent control over advanced computing power while reinforcing our leadership in AI and quantum.”

Arguably, the U.S. is taking a more aggressive and coordinated approach to building its computing infrastructure. Washington is aligning federal, state, and private-sector resources to ensure capacity is scaled quickly and domestically controlled, according to their announcement, in the coming weeks.

Recent executive orders have opened federal land for gigawatt-scale AI data centers, with the Departments of Defense and Energy identifying more than a dozen priority sites for fast-track construction. A White House task force now oversees permitting, cutting timelines through exemptions and priority reviews, while Congress advances legislation to expand grid capacity to handle AI’s energy demands. Canada’s current plan does not address these climate concerns and costs.

Private ventures are also set to contribute to growth, with projects like Stargate led by OpenAI, SoftBank, Oracle, and MGX planning billions in infrastructure spending. However, its initially announced goals have been scaled back.

By comparison, Canada’s sovereign computing strategy remains in its early stages, with much of its AI infrastructure still dependent on foreign ownership or operation. Canada also lacks a large-scale semiconductor manufacturing strategy, leaving it vulnerable to both pricing pressures and export restrictions from foreign suppliers like the U.S. and China.

Does “sovereign” serve the bigger picture?

Sovereignty is central to the government's AI strategy, but Canada’s record on using it to protect domestic ownership is mixed. Telecom is a clear example: appeals to national interest have justified a concentrated market, yet Canadians pay some of the highest mobile and internet prices in the G7.

Building infrastructure at home is not enough if it leads to high costs, limited competition, and barriers to access. Sovereignty should mean ensuring Canadians can actually use AI, benefit from it, and have a role in shaping how it is governed.

The real test is whether Canada can avoid repeating its history of digital infrastructure that undermined accessibility and affordability. Whether it leans toward the U.S.’s pro-industry stance or the EU’s stricter rules on public interest remains an open question.

Erick Espinosa is a Canadian journalist and digital storyteller with over a decade of experience creating content for leading news outlets. He specializes in travel, technology, and tourism and contributes to various publications, as well as hosting The Sociable's Brains Byte Back Podcast.