Canada’s AI strategy will protect women online. Not their jobs.
What this article covers
The World Economic Forum projects 92 million jobs displaced by 2030, and 170 million created. The jobs eliminated and the jobs created are not the same jobs. A marketing coordinator doesn’t become an AI engineer.
Why women are most at risk
Three reasons:
- Women are overrepresented in the jobs most exposed to AI,
- underrepresented in AI and STEM roles, and
- subject to AI systems built on biased data.
A 2025 ILO-NASK report found that if highly exposed jobs disappeared, two women would be displaced for every man.
Those jobs are administrative coordination, content production, customer support, bilingual communications. Work AI can now do well enough to replace people.
What about the trades?
Skilled trades are AI-proof for now. But few women make it in. Women make up 4.8% of construction apprentice registrations. Only a 2% completion rate. That’s not a skills problem. A YWCA Halifax study found more than 90% of women and gender-diverse tradespeople reported being harassed at work.
The jobs AI is eliminating are the ones women could get. The jobs AI can’t touch are the ones women struggle to get into.
Does Canada’s AI strategy protect women?
The strategy names women twice. The two mentions are not equal.
- Online violence
In its first pillar, the strategy names deepfakes as a form of sexual violence, particularly against women and children. It promises new online safety laws, privacy reform, and legal tools to fight them. Women appear here as people to be protected from harm, and the government knows how to legislate against a harm. - Economic displacement
The one labour commitment for women sits in the second pillar. The strategy’s answer is to accelerate AI adoption in the sectors where women are most at risk. Their word for it: softening transition shocks. And they will be using money already allocated to the existing Women’s Program.
Why one gets a law and the other doesn’t
Online violence has a clear perpetrator. Government knows how to write a law against that. And pairing women with children makes it a harm no one will oppose.
Economic displacement has a villain. But it’s the same industry this strategy is built to grow.
So one harm gets a bill. The other gets a process and a repurposed program. The strategy will defend women’s safety. Their economic future is another matter.
Worth noting: the strategy’s single largest line item is $700 million for data centres and AI infrastructure. When Ottawa wants to fund something, it funds it.
The bottom line
The global economy already loses more than $7 trillion a year to gender inequality at work. AI is set to widen that gap. Protecting women from deepfakes is necessary. It is not the same as protecting their place in the economy.
This isn’t about blame. Nobody in government set out to push women backward. But intent doesn’t change the outcome.
Women were already crowded into the jobs most at risk. They were already kept out of the ones that are safe. Now there’s a national plan that guards them from deepfakes and stays silent on their place in the economy.
A job is not just income. It’s independence and financial autonomy that took generations to win. When women lose work faster than men and the plan says little about it, it’s not just a paycheque at risk. It’s everything that paycheque made possible.
Resources
The data
- New ILO data confirm women face higher workplace risks from generative AI than men / International Labour Organization, March 2026. The primary source: 29% of jobs in female-dominated occupations are exposed to generative AI, versus 16% in male-dominated ones.
- Is AI changing the path to gender parity? / World Economic Forum, March 2026. Layers on the displacement risk, including the finding that women’s share of European tech roles is falling, not rising.
- We’re at risk of a two-tier AI economy if we don’t bridge the AI gender gap / Fortune, March 2026. Cites a Brookings analysis: of the workers least able to recover from AI job loss, 86% are women.
The Canadian picture
- The women most at risk from AI aren’t in the rooms designing it / BetaKit, April 2026. The sharpest Canadian framing: women here are nearly three times more at risk from automation than men, while Canada funds acceleration without equity.
- Canada’s labour protections aren’t ready for the age of AI / Policy Options (IRPP), March 2026. The structural argument — weak labour protection shifts the cost of AI displacement onto the workers least equipped to absorb it.
- Labour Force Survey, February 2026 / Statistics Canada. The baseline: women still earn $0.88 to the dollar and remain concentrated in part-time and clerical work, the exact footing AI is eroding.
Why the gap compounds
- Nearly half of women concerned about backlash from AI use at work / Talent Canada, March 2026. Canadian data: 36% of women feel confident using AI versus 52% of men, the confidence gap that turns into an opportunity gap.
- Women and AI: The Gender Gap in AI Adoption and Recognition / Lean In, 2026. Women get less credit and less manager support for using AI, so today’s small usage gap widens into tomorrow’s career gap.