Canada's new artificial intelligence strategy, AI for All , presents an ambitious vision for the country's future. Artificial intelligence, the federal government argues, can boost productivity, strengthen competitiveness and create opportunity across the economy.
But what happens when AI increasingly decides who gets access to those opportunities in the first place?
As migration scholars at the Bridging Divides initiative at Toronto Metropolitan University, we are researching how AI is reshaping immigrant access to employment. We see a growing intersection between Canada's immigration agenda and its AI agenda, both of which are essential to Canada's economic future. Yet these agendas are rarely discussed together.
As AI becomes increasingly embedded in hiring, the question is no longer just who Canada admits, but how opportunity is distributed once immigrants arrive.
The technologies shaping recruitment today may play an increasingly important role in determining whether Canada's skilled immigrants can fully contribute to the economy tomorrow.
Canada's reliance on immigration
Across the OECD, immigrants represent a growing share of the workforce . They are increasingly central to economic growth, demographic sustainability and innovation.
In Canada, immigrants accounted for four-fifths of labour force growth between 2016 and 2021 , underscoring Canada's continued reliance on immigration to sustain its workforce.
At the same time, Canada is rapidly embracing artificial intelligence. The use of AI across workplaces is expanding quickly, including in recruitment, hiring and workforce management. Recent evidence from Statistics Canada suggests that AI adoption among Canadian businesses has doubled in the past year .
These developments are often discussed separately. They should not be. The future success of one may increasingly depend on the other.
Immigrants are underemployed
Canada's immigration system is widely regarded as one of the world's most developed systems for selecting skilled migrants. Yet labour market integration remains one of the most persistent challenges facing immigrants.
Despite high levels of education and professional experience, immigrants continue to face higher rates of underemployment and over-qualification than Canadian-born workers. Statistics Canada reports that nearly one-third of recent immigrants with post-secondary education are overqualified for their jobs , compared with fewer than one in five Canadian-born workers.
Many continue to encounter barriers linked to foreign credential recognition, limited professional networks and demands for so-called " Canadian experience ." Research has long documented how internationally acquired skills and experience are often undervalued in Canadian labour markets .
These challenges are not new. What is new is the infrastructure through which they increasingly operate.
Algorithms as gatekeepers
Historically, access to employment was mediated largely through human institutions and judgments, from credential recognition bodies and hiring managers to recruiters and professional networks that determined whose skills were recognized and whose were overlooked. Today, many decisions occur earlier, within digital systems that evaluate, rank and filter applicants before a human recruiter reviews an application.
Across Canada and other OECD countries, recruitment is being transformed by applicant tracking systems, automated screening tools, predictive analytics and AI-enabled hiring platforms.
A growing body of research has raised concerns about how these systems can reproduce existing inequalities through the data and assumptions on which they rely. A study from Cornell University found that many claims about fairness in algorithmic hiring remain difficult to verify because hiring systems often inherit patterns embedded in historical recruitment data .
Similarly, Safiya Noble, a professor of gender studies and African American studies at the University of California, Los Angeles has shown how seemingly neutral digital systems can reproduce broader social inequalities in her book, Algorithms of Oppression: How Search Engines Reinforce Racism.
The OECD Employment Outlook warned back in 2023 that AI systems used in employment can raise significant concerns around transparency, explainability, accountability and discrimination, particularly for groups already facing labour market disadvantages.
AI is moving beyond simply providing advice to increasingly exercising authority in decisions that affect people's lives and livelihoods. Hiring is one of those decisions.
'Black box' screening systems
Preliminary findings from our recent Bridging Divides study, based on interviews with immigrants, employers and recruitment professionals across Canada, suggest many immigrants experience digital hiring systems as a "black box" that shapes outcomes but remains largely invisible to those affected.
One participant described repeatedly applying for positions that matched their qualifications:
"By the description itself, you felt that you are the right candidate, but somehow your resume is not picked up because of some reason that they said, like your resume is not matching the criteria."
The issue is not simply rejection. It is uncertainty. Applicants often do not know whether they were rejected because of qualifications, competition, employer preferences or automated screening systems.
Another participant explained:
"I applied for seven jobs without response. That's when I went to ChatGPT and researched why I might not be getting responses."
Many participants described learning to adapt their applications not for recruiters, but for algorithms. In effect, job seekers are increasingly required not only to demonstrate competence but also to make themselves legible to machines.
Equity in Canada's AI future
None of this means that AI is inherently discriminatory, nor does it mean employers should abandon digital hiring technologies. Used responsibly, these tools can improve efficiency and help employers identify qualified candidates.
But Canada has spent decades refining how it selects skilled immigrants. Less attention is being paid to the technologies increasingly shaping whether those skills are recognized after arrival.
As AI becomes embedded in recruitment, we need to consider how opportunity is governed once immigrants enter the labour market. If immigration policy selects skilled immigrants, AI is increasingly shaping whether they are seen as skilled workers.
Canada's future depends on both immigration and artificial intelligence. Ensuring that these two ambitions reinforce rather than undermine one another may become one of the defining policy challenges of the AI era.
Hari KC is Research Fellow with the Bridging Divides initiative at Toronto Metropolitan University
Rupa Banerjee receives funding from the Canada Research Chairs Program and the Canada First Research Excellence Fund.
