Most conversations about AI agents are about what they do inside a company. This one is about who the company hires to work next to them. In May 2026, KPMG Canada put a number on it: 59% of surveyed Canadian firms say AI agents have already changed how they hire entry-level workers.
If you run an Ontario business, that shapes who's in your talent pool. If you're a newcomer looking for a first Canadian job, it shapes the postings you'll see, and how they get read. Here's what the data supports, and where the headlines run ahead of it.
Key Takeaways
- 59% of Canadian firms say AI agents changed entry-level hiring, and 63% say the same for experienced hires (KPMG Canada, May 2026).
- The survey covered only firms with $50M+ in revenue, so it says little about small Ontario employers.
- "Changed" doesn't mean "cut." Part of the entry-level squeeze predates ChatGPT, and the job market is soft anyway.
- Since January 1, 2026, Ontario employers with 25+ staff must disclose AI screening in public job postings.
- Smaller employers can treat the squeeze as a talent opening, not a threat.
What Did KPMG's 2026 Survey Actually Find?
In a survey of 306 Canadian executives fielded February 19 to March 27, 2026, KPMG Canada found 77% already use AI agents for tasks like sharing knowledge between departments, and 66% are moving toward a fully integrated human-AI workforce (KPMG Canada, May 2026). Hiring is where the shift shows up: 59% say agents changed entry-level hiring, and 63% say they changed how experienced talent is hired.
Notice the wording, though. KPMG reports that hiring has changed, not that it has shrunk. The release doesn't publish headcount or job-cut figures. Any headline that turns 59% into "59% of firms are replacing juniors with AI" is claiming more than the survey says.
There's a bigger limit, and most coverage skips it. Every one of the 306 firms has at least $50 million in annual revenue, and about a quarter earn more than $500 million. A ten- or forty-person Ontario company isn't in the sample at all. So read the survey as a signal from larger employers, who set the tone for the junior-hiring market that smaller firms and newcomers then compete in.
Is AI Really the Reason Entry-Level Jobs Are Harder to Find?
Partly, but nobody can yet say how much. As of August 2026, Statistics Canada's Labour Force Survey put youth (15 to 24) unemployment at 12.9% against 6.4% overall, with wage growth at 2.0%, the slowest since late 2017 outside the pandemic years (Statistics Canada, Labour Force Survey, August 2026). That's a weak market for juniors whatever AI is doing.
The tech sector gives the clearest hint of an AI effect, and a warning against over-reading it. Indeed Hiring Lab found Canadian tech postings were down 19% from early 2020 by August 2025. Junior and standard-level postings fell 25% while senior and manager postings rose 5% (Indeed Hiring Lab, Canadian Tech Hiring Freeze Continues, August 2025). Yet roughly half of that decline happened before ChatGPT launched.
A staffing-industry paper adds employer sentiment. Express Employment Professionals found 81% of hiring managers say entry-level roles now demand more skills, and 45% see using AI for entry-level tasks as more efficient than hiring and training someone (reported by HRD Canada, May 2026). That's a staffing firm's survey, not a government statistic, so treat it as sentiment.
Put those together and the honest read is a squeeze with several causes: a soft economy, tighter skill expectations, and AI giving employers a new option for routine work. If you're a job-seeker, that matters. It means the practical fix is rarely "AI-proof yourself" and usually "show a skill an employer can't get from a tool."
Why Are Canadian Workers Resisting AI Agents?
In KPMG's 2026 survey, 31% of Canadian firms report employee resistance to AI agents, nearly double the 16% global figure. The leading reasons are trust, ethics and transparency concerns (51%), job security (39%), and low confidence in AI skills (38%) (KPMG Canada, The Agentic Shift, May 2026).
Why does this belong in a hiring post? Because the same trust gap applies to applicants. People sending resumes into an unseen screening system have the same questions about how decisions get made, and Ontario has now written a rule about it.
What Does Ontario's AI Job-Posting Rule Mean in Practice?
Since January 1, 2026, Ontario employers with 25 or more employees must say in a publicly advertised job posting, and in the application form, if AI is used to screen, assess or select applicants (Alexander Holburn Beaudin + Lang LLP via Mondaq, January 2026). Internal postings and roles performed outside Ontario are excluded.
The same rules reach beyond AI. Postings must include an expected pay range, the range can't span more than $50,000, and roles paying over $200,000 are exempt. Postings and application forms also can't contain a Canadian experience requirement, and employers must keep copies for three years after a posting comes down (Osler, AI in Hiring: Ontario Employers Grappling With New Job Posting Disclosure Requirement).
There's an open question employers should know about. The regulation defines AI broadly, as a machine-based system that infers outputs from inputs, but the Ministry hasn't defined "screen," "assess" or "select," or said how much detail a disclosure needs. Does a resume-parsing feature in your applicant tracking system count? Counsel hasn't settled that, so check your own tools rather than assuming you're outside the rule. Because AI tools handle personal data, our PIPEDA compliance guide covers the privacy side.
What Should Ontario Employers Do About It?
Treat the squeeze as a decision, not a trend to wait out. Larger firms are changing their junior hiring while smaller employers sit outside the data. That leaves a gap, and what you do with it is a design question.
Watts Group works with newcomer founders across Ontario, and the pattern we see is consistent. The owners who get the most from early-career hires aren't the ones with the most automation. They're the ones who write down what a new hire should be able to do in 90 days, then pair them with someone who can check the work. Here's a practical sequence.
Audit Which Entry-Level Tasks Are Routine
List the tasks a junior would do. Data entry, first-draft replies and scheduling are candidates for an AI tool. Client follow-up, checking and judgment calls usually aren't. Hire for the second list.
Check Whether You're Covered by the AI Rule
At 25 or more employees, review every tool that ranks, filters or scores applicants. If you're under 25, disclosing AI use anyway is a low-cost trust signal to candidates.
Clean Up Your Postings
Add a pay range, remove any "Canadian experience required" wording, and keep a copy of each posting. These are the same rules in the same section of the Employment Standards Act.
Keep a Human Reviewing Every Shortlist
A tool can sort applicants, but someone on your team should read the ones it rejects. KPMG's own data shows trust is the main barrier to AI use, so being able to explain a hiring decision matters.
Build a Real First 90 Days
Only 3% of Canadian firms in KPMG's survey report measurable AI returns. Don't replace a trainable hire with a tool you can't yet measure. Train people on how to work with AI, and track what improves.
What Should Newcomer Job-Seekers Do Differently?
Read the posting before you apply. Under Ontario's 2026 rules, a compliant public posting from a 25-plus employer tells you whether AI is screening, what the pay range is, and that Canadian experience can't be required. A posting that skips all three isn't automatically illegal, since small employers are exempt from the AI line, but it tells you something about the employer.
Then write for both readers. Mirror the posting's own job-title and skill wording, use a plain single-column layout with standard headings, and put numbers on results. No Canadian source confirms how any specific tool ranks resumes, so treat this as sensible practice rather than a formula. And don't skip people: a referral, a LinkedIn message to the hiring manager, or a volunteer placement still carries weight when the first filter is a machine.
Is the goal to beat the system? Not really. It's to be easy to read for a screening tool and easy to remember for the person who decides. Foreign credentials, a clear Canadian-market story and one specific project you can describe in two sentences help with both.
Frequently Asked Questions: AI Agents and Hiring in Canada
Are AI agents taking entry-level jobs in Canada?
Not provably. KPMG Canada's 2026 survey found 59% of Canadian firms say AI agents have changed how they hire entry-level workers, but "changed" isn't "cut." Indeed Hiring Lab found about half of the drop in Canadian tech postings came before ChatGPT launched, so a weak job market is also at work.
Does Ontario require employers to say if they use AI in hiring?
Yes, for larger employers. Since January 1, 2026, Ontario employers with 25 or more employees must state in a publicly advertised job posting, and in the application form, if AI is used to screen, assess or select applicants. Smaller employers and internal postings are not covered.
Can Ontario employers still require Canadian experience?
Not in a publicly advertised job posting. Ontario's 2026 job-posting rules prohibit any Canadian experience requirement in postings and application forms, and require a pay range in most postings. That's useful for newcomers, who can spot non-compliant postings quickly.
How should a newcomer write a resume that AI screening tools can read?
Mirror the posting's own job-title and skill wording, use a plain single-column format with standard headings, and quantify results. No Canadian source confirms how any specific screening tool ranks resumes, so treat this as sensible practice, not a guaranteed formula.
Is it worth a small Ontario business hiring junior talent right now?
Often yes. KPMG's survey covered only firms with $50M or more in revenue, so small firms aren't in the data. If larger employers narrow their junior intake, a smaller employer that trains well can access capable newcomer talent with less competition, provided it plans supervision and clear tasks.
AI agents are changing hiring in Canada, and the clearest evidence is how larger firms say their process has shifted. What the data doesn't show is a collapse in junior jobs caused by AI alone. For an Ontario employer, the practical move is to audit your tasks, clean up your postings and keep a person accountable for every hire. For a newcomer, it's to read postings closely and make your skills easy to see. If you're also weighing where AI belongs inside your own business, our guide on whether you actually need an AI agent is the place to start.
