Canada sits 44th of 47 countries on AI training and literacy while business adoption triples, and the order of those two facts is the problem.
Canada’s weakest AI number is training, and it is not close. KPMG International and the University of Melbourne surveyed more than 48,000 people across 47 countries, including 1,025 Canadians, and placed Canada 44th of 47 on AI training and literacy and 28th of 30 among advanced economies. Just 24% of Canadians reported receiving AI training, against 39% globally. Adoption in Canadian firms kept climbing anyway. That is how a company ends up with tools in the building and nobody trained to use them.
Key points
- Canada ranked 44th of 47 countries on AI training and literacy in the KPMG and University of Melbourne global study published June 23, 2025.
- 24% of Canadians reported AI training, against a global average of 39%.
- 38% of Canadians claimed moderate or high AI knowledge, against 52% globally.
- Statistics Canada found 32.0% of Canadian businesses using AI provided training to existing employees, and 21.6% trained existing senior leaders.
- Training investment splits sharply by size: 68.1% of firms with 100 or more employees trained existing staff, against 24.0% of micro-businesses.
What the AI training numbers say about Canada
The KPMG and University of Melbourne study is a survey of people, not firms, and the Canadian result is consistent across every measure it takes. Training received, 24% against 39% globally. Self-assessed AI knowledge at moderate or high, 38% against 52%. Confidence in using AI effectively, 47% against 60%.
| Measure | Result |
|---|---|
| Canada against the global average | |
| Received AI training, Canada | 24% |
| Received AI training, global | 39% |
| Moderate or high AI knowledge, Canada | 38% |
| Moderate or high AI knowledge, global | 52% |
| Confident using AI, Canada | 47% |
| Confident using AI, global | 60% |
Trust follows the same curve. Canada placed 42nd of 47 on willingness to trust AI systems, and 46% of Canadians said the risks outweigh the benefits, against 32% globally.
Read those together. The usual reading is that Canadians are cautious by temperament. A less flattering reading fits the numbers better: people distrust a tool nobody showed them how to use. That is an analysis, not a finding, and it would fall apart if trained Canadians turned out to trust AI no more than untrained ones.
Why it matters in Canada
Because adoption did not wait for the training. Statistics Canada reported that 19.2% of businesses used AI to produce goods or deliver services in the second quarter of 2026, up from 6.1% two years earlier.
The same release holds the number an owner should lose sleep over. Among businesses using AI, 32.0% provided training to existing employees and 21.6% trained existing senior leaders. So roughly two thirds of AI-using Canadian firms rolled out a technology and trained nobody already on payroll.
Size decides it. 68.1% of firms with 100 or more employees trained existing staff, against 24.0% of micro-businesses. The companies with the least slack to absorb a failed rollout are the ones going in without preparation.
What untrained adoption costs
Three predictable bills arrive.
Staff use the tool badly, conclude it does not work, and stop without telling anyone. You keep paying the licence. Second, untrained users reach for whatever is easiest, which is a personal account on a consumer app with company information pasted into it. Third, leaders who never trained cannot tell a good AI proposal from a bad one, which is how a business buys the wrong platform at the wrong price.
None of that shows up as a training line in the budget. It shows up as churn, as a breach, and as a renewal nobody can justify.
What should leaders do next
- Train the leadership team first. 21.6% is the number to beat. The people approving AI spending should not be the least prepared in the room.
- Run two hours, not two days. Approved tools, what data never goes in, three real tasks from your own business, and one worked example each.
- Pick the tasks from work people already do. Generic prompt-writing courses teach a skill nobody uses on Monday morning.
- Repeat it quarterly. The tools change faster than any annual training cycle.
- Measure something afterwards. Share of staff using the approved tool weekly is enough, and it beats a completion certificate.
Related: before cutting junior roles in the name of AI, read StatCan’s findings on youth hiring and AI exposure.
FAQ
Where does Canada rank on AI literacy?
44th of 47 countries on AI training and literacy, and 28th of 30 advanced economies, in the KPMG and University of Melbourne global study published in June 2025.
How many Canadian businesses train their staff on AI?
Statistics Canada found 32.0% of AI-using businesses trained existing employees and 21.6% trained existing senior leaders in the second quarter of 2026.
How much AI training does a small business actually need?
Two hours covering approved tools, prohibited data and three real tasks, repeated quarterly, puts a small firm ahead of most of its peers.
Related reading
- AI Implementation in Canada Needs Repeatable Workflows
- AI Tool Fatigue and the Business AI Switchboard
- AI Transparency in Canada Needs a Practical Company Record
Disclosure
The author has no relevant financial, advisory, or board relationships with any party named in this brief.
More in the same thread on our AI for business in Canada reference.
