Is Any Degree Future-Proof? What Canadian Data Actually Says About AI and Careers
· 4 min read
Every parent of a teen has heard some version of it: don't bother with that field, AI will do the job. It's a fair worry. But most of the confident advice going around — learn to code, don't learn to code, only the trades are safe — runs ahead of the evidence. Statistics Canada has been tracking this closely since 2024, and what the data shows is more useful than any list of "safe" majors.
Exposure isn't the same as replacement
Statistics Canada sorts jobs on two measures: how much of the work AI could touch (exposure), and whether AI is more likely to help the person doing it or substitute for them (complementarity).
By that measure, 29% of Canadian workers are in jobs highly exposed to AI and highly complementary with it, 31% are in highly exposed jobs where AI is more likely to substitute, and 40% are in jobs with low exposure.
The first group includes doctors, nurses, teachers and engineers. The second includes data entry and office support roles — but also some jobs people assume are safe, like web designers, database analysts, accountants and software engineers. (StatCan notes real uncertainty on that last one: someone has to build and maintain the AI.)
What the "AI helps, not replaces" jobs have in common
StatCan's reasoning for which jobs AI is more likely to complement comes down to a few traits: responsibility for someone's health or safety, lots of face-to-face judgment, work in complex physical settings, and mistakes with serious consequences. A useful gut check for any program your teen is considering: when this job goes wrong, does a person have to answer for it?
Three years in, here's what actually happened
StatCan's early-evidence study covering November 2022 to December 2025 found employment kept growing across all three groups — no clear collapse in the most-exposed jobs.
The sharper divide was by education. Workers with a bachelor's degree or higher saw job gains of 10–20%, and trades certificate holders gained too. Workers with a high school diploma or less — or with incomplete post-secondary — saw little to no growth.
The warning sign is at the entry door, especially in tech. Coding-heavy jobs grew overall, but employment for coders under 30 stayed flat while it rose almost 30% for those aged 30–49, and postings for coding jobs needing three years' experience or less fell by about 60%. StatCan cautions this may partly be a correction after the pandemic hiring boom. Either way, the practical lesson is the same: the first job in tech is harder to land than it was.
Where the skilled trades fit
Every certified trade StatCan examined — plumbers, carpenters, welders, machinists, heavy-duty mechanics and more — landed in the low AI-exposure group. The catch: about 20% of trades workers are at high risk of automation (machines and robotics, not chatbots), versus 13% in other jobs. Trades aren't untouched by technology, but the change looks different — and it's slower and easier to see coming. (Related: why so many teens overlook the trades.)
What this means for choosing a program
Choose the work, not the label. Roles with real accountability — nursing, teaching, engineering, licensed trades — sit in the stronger position regardless of what the degree is called.
If tech is the pull, go — but plan for experience. Coding-intensive jobs still paid about $54 an hour on average in 2025, well above other jobs. The difference now is that experience before graduation matters more, which makes co-op programs worth prioritizing.
Finish the credential. The weakest job growth was among people without a completed post-secondary credential. Unfinished isn't neutral.
Fit still beats forecasting. Nobody knows exactly what 2035 looks like. A teen in a field that genuinely suits them is better positioned to adapt than one who picked a "safe" major they don't care about.
How Ypath helps
Ypath matches your teen to career paths based on who they actually are, then shows real job titles in each path with Government of Canada Job Bank wages and Ontario outlook ratings — sortable by income, free — so you're working from real numbers instead of headlines.
FAQ
Is computer science still worth it?
The data says yes, with a caveat. Coding-intensive jobs grew and still pay well above average, but entry-level hiring has tightened sharply. Programs with co-op or strong work-placement options matter more than they used to.
Are the skilled trades AI-proof?
They have among the lowest exposure to AI of any occupations in Canada. They do face a somewhat higher risk of automation from machinery and robotics, which tends to change jobs gradually rather than replace them outright.
Which degrees are the most future-proof?
No degree is guaranteed. The strongest position is in work AI is more likely to assist than replace — jobs involving responsibility for people's health or safety, face-to-face judgment, and hands-on work — such as nursing, teaching, engineering and licensed trades.
Sources
- Statistics Canada, Experimental Estimates of Potential Artificial Intelligence Occupational Exposure in Canada (Mehdi & Morissette, 2024) — https://www150.statcan.gc.ca/n1/daily-quotidien/240903/dq240903b-eng.htm
- Statistics Canada, Canadian employment trends in the era of generative artificial intelligence: Early evidence (Mehdi & Frenette, January 2026) — https://www150.statcan.gc.ca/n1/pub/36-28-0001/2026001/article/00003-eng.htm
- Statistics Canada, Potential occupational exposure to artificial intelligence and automation among certified journeypersons in Canada (Leanage & Mehdi, January 2026) — https://www150.statcan.gc.ca/n1/pub/36-28-0001/2026001/article/00001-eng.htm