More than one-third of Canadian workers used generative artificial intelligence tools at work in the 12 months to March 2026, with adoption highest in management, scientific and other occupations with greater potential exposure to AI, according to new Statistics Canada data.
The agency's first results from a regular series examining AI use by Canadian workers show that 35.9 per cent of workers aged 15 to 69 used generative AI as part of their main job or business during the period.
The findings are based on supplementary questions to the Labour Force Survey collected in March 2026.
Use varies sharply by industry and occupation
Overall, 41.6 per cent of workers reported using at least one AI or automation technology in their main job or business during the previous 12 months. Generative AI tools were the most commonly reported technology.
Use was highest in professional, scientific and technical services, where 65.6 per cent of workers reported using generative AI. That was followed by finance, insurance, real estate, rental and leasing at 59.2 per cent and educational services at 53.0 per cent.
The lowest rates were reported in accommodation and food services at 16.3 per cent, agriculture at 17.5 per cent and transportation and warehousing at 21.1 per cent.
By broad occupational group, use was highest among management workers, at 75.1 per cent, and workers in natural and applied sciences, at 67.5 per cent.
Use was lowest among trades, transport and equipment operators, at 14.7 per cent, and natural resource, agriculture and related occupations, at 17.0 per cent.
Exposure to AI linked to higher use
Statistics Canada grouped occupations according to their potential exposure to AI and the degree to which AI could complement the work performed.
The three categories are high exposure and high complementarity, or HEHC; high exposure and low complementarity, or HELC; and low exposure, or LE.
In March, 31.2 per cent of workers aged 15 to 69 were employed in HEHC occupations, 29.6 per cent were in HELC occupations and 39.3 per cent were in LE occupations.
Generative AI use was highest among workers in HEHC occupations, with 53.8 per cent reporting use at work. That was followed by 45.9 per cent among workers in HELC occupations and 14.2 per cent among workers in LE occupations.
The release says HEHC occupations, including doctors, nurses, teachers and engineers, involve tasks with greater potential complementarity with AI and might therefore be more likely to benefit from the technology.
HELC occupations, including retail sales, office support, software development and accounting, may be more susceptible to task replacement by AI. LE occupations, such as skilled trades, service jobs and first responders, may be less likely to be affected by AI-related transformation.
Mikhail Nilov photo
Age and gender differences emerge
Use of generative AI also varied by age within the occupational categories.
Among workers in HEHC occupations, 56.9 per cent of core-aged workers between 25 and 54 used generative AI at work, compared with 45.3 per cent of workers aged 55 and older and 39.1 per cent of youth aged 15 to 24.
In HELC occupations, 51.9 per cent of core-aged workers used generative AI, compared with 33.2 per cent of young workers and 31.5 per cent of older workers.
Use was lower across LE occupations, particularly among older workers. Just 6.7 per cent of workers aged 55 and older in those occupations reported using generative AI, compared with 15.3 per cent of youth and 16.3 per cent of core-aged workers.
Men were generally more likely than women to report using generative AI within the HEHC and HELC categories.
In HEHC occupations, 57.0 per cent of men reported using the tools, compared with 50.9 per cent of women. In HELC occupations, the corresponding figures were 52.9 per cent for men and 41.2 per cent for women.
There was virtually no gender difference in LE occupations, where 14.1 per cent of men and 14.4 per cent of women reported using generative AI at work.
Most workers use AI for some tasks, rather than across their jobs
Among workers who used generative AI during the previous 12 months, nearly two-thirds, or 63.5 per cent, reported using the tools for some but not most tasks.
Another 24.9 per cent reported minimal use, meaning they used generative AI for almost no tasks.
Broader use remained uncommon. About 8.0 per cent of users said they used generative AI for most tasks, while 3.6 per cent used it for almost all tasks, for a combined total of 11.6 per cent.
Moderate use was the most common pattern across all three occupational categories. It was reported by 67.6 per cent of users in HEHC occupations, 63.6 per cent in HELC occupations and 50.7 per cent in LE occupations.
Broad use was most common among HELC workers, at 15.4 per cent, compared with 9.2 per cent among HEHC workers and 9.9 per cent among LE workers.
Daily and weekly use common among those who use the tools
Among workers who used generative AI at work, 31.4 per cent reported using the tools daily and 38.3 per cent used them a few times per week.
Another 21.7 per cent reported using AI a few times per month, while 8.7 per cent used it a few times a year.
Daily use was particularly common among workers in natural and applied sciences, where 45.6 per cent of users reported using generative AI every day.
By comparison, daily use was reported by 18.6 per cent of users in manufacturing and utilities and 18.2 per cent of users in natural resources, agriculture and related occupations.
Applicability remains the main reason for not using AI
Despite widespread awareness of generative AI, most workers did not use the tools at work during the previous 12 months.
About 93.4 per cent of workers said they were aware of generative AI tools. Among those workers, 51.5 per cent were familiar with how the tools could be applied to their current work, including 15.0 per cent who were very familiar.
Another 11.1 per cent said they were not familiar with how generative AI could be applied to their work, while 37.4 per cent said they were familiar with the tools but did not believe they were applicable to their work.
Among workers who had not used generative AI — 64.1 per cent of all workers — the most common reason was that the tools were not applicable to their work. That reason was cited by 56.0 per cent of non-users.
The share was higher among private-sector employees, at 59.8 per cent, compared with 47.7 per cent among public-sector employees and 48.6 per cent among self-employed workers.
One in five non-users, or 21.6 per cent, said they had no interest in using generative AI. That figure was higher among self-employed workers, at 26.9 per cent, than among public-sector employees at 21.4 per cent and private-sector employees at 20.8 per cent.
Security, privacy, environmental or ethical concerns were cited by 9.8 per cent of non-users. Public-sector employees were more likely to report those concerns, at 16.9 per cent, compared with 7.4 per cent among private-sector employees and 11.1 per cent among self-employed workers.
A lack of skills or knowledge was cited by 5.8 per cent of non-users, while 5.0 per cent said company or organizational policies limited their use of generative AI tools.
Public-sector use higher overall
Generative AI use was reported by 41.2 per cent of public-sector employees, compared with 33.4 per cent of private-sector employees and 39.6 per cent of self-employed workers.
Statistics Canada said part of the difference reflected the higher share of occupations with high potential exposure to AI in the public sector.
Within HEHC occupations, use was similar between private- and public-sector workers, at 54.9 per cent and 54.5 per cent respectively.
The gap was also small in HELC occupations, where 45.8 per cent of private-sector workers and 43.1 per cent of public-sector workers reported using generative AI.
The results mark the first release in Statistics Canada's regular series on AI use by Canadian workers and provide a March 2026 snapshot of how widely the technology was being used across workplaces and occupations.