Which ecommerce jobs AI reaches first: Canadian evidence to 2026

Nine jobs, one parcel

An online order becomes work the moment it is paid. Someone receives the stock and checks it against the documents. Someone stores it and keeps the count. Someone picks it, packs it and prints the label. A dispatcher builds the route, a driver runs it, and a courier or delivery driver hands the parcel over. Statistics Canada classifies that chain into nine occupations: shippers and receivers, storekeepers and partspersons, dispatchers, supply chain supervisors, logistics coordinators, material handlers, delivery service drivers, transport truck drivers, and couriers and messengers.

In mayo de 2021 those nine occupations employed 923,250 people, 5.3% of the 17,321,700 Canadians employed that month. Truck drivers were the largest group at 342,340, material handlers next at 203,260, then shippers and receivers at 110,105. Delivery service drivers were 90,490, couriers and messengers 32,095 and dispatchers 28,900.

That count is five years old, and there is no newer one. The Labour Force Survey publishes occupations only down to sub-major groups, so no monthly or annual figure exists for shippers and receivers, or for delivery drivers, on their own. This brief uses the Census for the size of the work and the survey groups, which hold other jobs too, for its direction.

What exposed means, and which of the nine it describes

Statistics Canada's measure asks two questions of every occupation. How much do the abilities of current AI overlap the tasks the job is made of, and does a person still have to be in the loop when it does. An occupation above the median on the first question is high exposure. One above the median on the second is high complementarity. The estimates are built on the May 2021 Census and on task descriptions from O*NET.

On that measure, 31% of Canadian employees were in high-exposure, low-complementarity jobs in May 2021, 29% in high-exposure, high-complementarity jobs and 40% in low-exposure jobs.

Applied to the parcel, the line falls between the desk and the floor. Office support and co-ordination occupations, the group that holds shippers and receivers, storekeepers, dispatchers and schedulers, are 76% high exposure and low complementarity and 24% low exposure. Administrative occupations in finance, insurance and business, which hold logistics coordinators, are 82% high exposure and low complementarity. Transport and heavy equipment operators and servicers, the group holding truck drivers, delivery drivers and material handlers, are 100% low exposure. Support occupations in sales and service are 99% low exposure.

By industry the same shape appears. Transportation and warehousing is 19% high exposure, low complementarity and 66% low exposure. Retail trade is 37% and 40%. Wholesale trade has 33% in the first group.

A second index reaches the same ordering by a different route. The International Labour Organization scores each occupation's tasks against generative AI from 0 to 1: data entry clerks 0.7, clearing and forwarding agents 0.57, transport clerks 0.5, production clerks 0.44, messengers and package deliverers 0.37, stock clerks 0.37, heavy truck drivers 0.24, freight handlers 0.14. The exposed tasks are the document, the record and the schedule. The unexposed tasks are the lift, the drive and the hand-over. Both indices say this. Neither says when.

What the usage data says the exposed tasks are

Anthropic's Economic Index reads the tasks people actually bring to its models. In early 2025, about 36% of occupations saw AI use in at least a quarter of their tasks and about 4% in three quarters, and use leaned to augmentation, 57% against 43% automation. By August 2025 automation exceeded augmentation on Claude.ai for the first time, as directive requests rose from 27% to 39% of conversations, and business traffic through the API ran 77% automation. By November 2025 augmentation was back to 52% against 45%, while office and administrative support tasks had risen to 13% of business transcripts, described as email management, document processing, customer relationship management and scheduling.

Canadian businesses report the same shape from the other side. 19.2% of all businesses used AI to produce goods or deliver services in the year to el segundo trimestre de 2026, a share that has tripled from 6.1% in the second quarter of 2024. Transportation and warehousing was at 9.8%, wholesale trade at 7.9%, retail trade at 19.2% and finance and insurance at 40.4%. Among transportation and warehousing businesses that used AI, 52.2% used virtual agents or chat bots, 33.0% data analytics, 22.2% large language models, 18.2% decision-making systems and 3.6% robotic process automation. Among retailers that used AI, data analytics led at 46.4%, then chat bots at 40.0% and marketing automation at 29.4%.

Workers say the same. Generative AI use at work rose from 17% of Canadian workers in September 2024 to 30% in July 2025 and 35.9% in March 2026. In transportation and warehousing it was 21.1% in March 2026; among trades, transport and equipment operators, 14.7%. Use ran at 53.8% in high-exposure, high-complementarity occupations and 14.2% in low-exposure ones. Of those who used the tools, 11.6% used them for most or almost all of their tasks.

None of these sources counts warehouse or delivery workers on their own. The closest cuts are the industry and the broad occupation group, and both put the floor near the bottom of the table.

Has the headcount moved

Statistics Canada answered this in January 2026 with the Labour Force Survey. From November 2022 to December 2025, employment grew at similar rates whatever an occupation's exposure and complementarity, with no persistent decline in high-exposure, low-complementarity jobs. Employment for men in those jobs was about 10% higher in December 2025 than in November 2022; for women, about 5% higher. Growth was little to none for employees with a high school diploma or less, in every exposure group, and the report notes that data entry and clerical occupations have decreased over time. Among businesses that adopted AI, about 6% reported reducing employment because of it in each of the last two survey periods.

The survey groups this brief can see moved the same way. Administrative and financial support and supply chain logistics occupations, the group holding shippers, receivers, storekeepers and dispatchers, went from 801.1 thousand in 2019 to 819.7 thousand in 2025, up 2.3%. Administrative and transportation logistics occupations rose 15.0%. Helpers, labourers and other transport drivers, the group holding material handlers and delivery drivers, rose 10.7%, from 546.4 thousand to 604.7 thousand. Mail and message distribution rose 12.5%, general trades, which include truck drivers, 6.5%, and all occupations 10.3%. By industry, transportation and warehousing employment rose 5.2% and retail trade 0.4%.

Three cautions. The supply chain logistics group fell from 981.7 thousand in 2014 to 820.0 thousand in 2016 with no stated reason in the table, so the comparison starts in 2019. Each group holds many jobs that have nothing to do with parcels; a clerk in a hospital counts in the same group as a shipper. And annual averages through 2025 cannot show anything that began in 2026.

What employers expect, labelled as expectation

The World Economic Forum's Future of Jobs Report 2025 surveyed employers in January 2025. They projected 170 million jobs created and 92 million displaced by 2030, a net gain of 78 million, or 7%. Light truck or delivery services drivers appear among the fifteen fastest-growing roles. Material-recording and stock-keeping clerks, data entry clerks and postal service clerks appear among the fastest-declining. Supply chain and logistics specialists and e-commerce specialists appear among the roles adding the most jobs. In supply chain and transportation, 94% of employers expected AI and big data skills to grow in importance by 2030.

These are expectations, not measurements. They point the same way as the Canadian exposure data: the clerk's desk changes first, and the driver's seat is being hired for.

How to read these numbers

For a merchant choosing a fulfillment or delivery partner, the exposed tasks are the documents, the dispatch records and the customer messages. Ask where those sit in the provider's operation and what system does them. The physical work is not what the indices describe.

For someone in the work, the Canadian data through December 2025 shows no headcount effect in exposed occupations. The change so far is in what the office tasks look like, and the workers using the tools are in the office jobs, not on the floor.

What the series cannot say: it has no count of any single parcel occupation after May 2021. It cannot attribute a change in headcount to AI. The exposure indices measure the overlap between tasks and abilities, not adoption and not intent.

Método

Occupation counts are Statistics Canada Table 98-10-0449-01 (Census of Population 2021, NOC 2021 unit groups, employed, randomly rounded to a multiple of 5). Labour Force Survey annual averages are Table 14-10-0416-01 (sub-major occupation groups) and Table 14-10-0023-01 (industries), persons in thousands, released 9 January 2026. Business AI use is Table 33-10-1167-01 (Canadian Survey on Business Conditions, second quarter of 2026, released 27 May 2026). Exposure shares are quoted from Mehdi and Morissette (2024) and from ILO Working Paper 140 (2025); worker-side use from Statistics Canada 75-006-X (June 2026) and The Daily of 30 July 2026; usage patterns from the Anthropic Economic Index reports of February 2025, September 2025 and January 2026; employer expectations from the WEF Future of Jobs Report 2025. Percentage changes are computed from the table values. Nothing is interpolated or forecast. Tables were retrieved through the Web Data Service on 13 September 2026.

Fuentes

1. Statistics Canada. Table 14-10-0416-01 Labour force characteristics by occupation, annual (NOC 2021 sub-major groups, employment, persons in thousands). Released 9 January 2026, accessed 13 September 2026. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1410041601 (se abre en una pestaña nueva)
2. Statistics Canada. Table 14-10-0023-01 Labour force characteristics by industry, annual. Released 9 January 2026, accessed 13 September 2026. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1410002301 (se abre en una pestaña nueva)
3. Statistics Canada. Table 98-10-0449-01 Occupation unit group by labour force status, Census of Population 2021. Released 30 November 2022, accessed 13 September 2026. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=9810044901 (se abre en una pestaña nueva)
4. Statistics Canada. Table 33-10-1167-01 Use of artificial intelligence by businesses and organizations in producing goods or delivering services over the last 12 months, second quarter of 2026. Released 27 May 2026, and "Analysis on artificial intelligence use by businesses in Canada, second quarter of 2026", 11-621-M No. 2026010, 11 June 2026. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=3310116701 (se abre en una pestaña nueva)
5. Mehdi, T. and Morissette, R. (2024). Experimental Estimates of Potential Artificial Intelligence Occupational Exposure in Canada. Statistics Canada, Analytical Studies Branch Research Paper 11F0019M No. 2024005, 3 September 2024. Chart 1 and Chart 2 data tables. https://www150.statcan.gc.ca/n1/pub/11f0019m/11f0019m2024005-eng.htm (se abre en una pestaña nueva)
6. Statistics Canada. "Exposure to artificial intelligence in Canadian jobs: Experimental estimates", Economic and Social Reports, 25 September 2024. https://www150.statcan.gc.ca/n1/pub/36-28-0001/2024009/article/00004-eng.htm (se abre en una pestaña nueva)
7. Statistics Canada. "Canadian employment trends in the era of generative artificial intelligence: Early evidence", Economic and Social Reports, 28 January 2026. https://www150.statcan.gc.ca/n1/pub/36-28-0001/2026001/article/00003-eng.htm (se abre en una pestaña nueva)
8. Statistics Canada. "Workplace artificial intelligence use: A profile of sociodemographic and job characteristics", Insights on Canadian Society, 75-006-X, 17 June 2026. https://www150.statcan.gc.ca/n1/pub/75-006-x/2026001/article/00007-eng.htm (se abre en una pestaña nueva)
9. Statistics Canada. The Daily, "Use of generative artificial intelligence tools among Canadian workers, March 2026", 30 July 2026. https://www150.statcan.gc.ca/n1/daily-quotidien/260730/dq260730b-eng.htm (se abre en una pestaña nueva)
10. Anthropic. The Anthropic Economic Index: "Introducing the Anthropic Economic Index" (10 February 2025), "Uneven AI adoption" (15 September 2025) and "Economic primitives" (15 January 2026). https://www.anthropic.com/economic-index (se abre en una pestaña nueva)
11. Gmyrek, P., Berg, J. et al. (2025). Generative AI and Jobs: A Refined Global Index of Occupational Exposure. ILO Working Paper 140, May 2025, Table A1. https://www.ilo.org/publications/working-papers/generative-ai-and-jobs-refined-global-index-occupational-exposure (se abre en una pestaña nueva)
12. World Economic Forum. The Future of Jobs Report 2025, January 2025. Figures 2.2, 2.3 and 3.5. https://www.weforum.org/publications/the-future-of-jobs-report-2025/ (se abre en una pestaña nueva)

Vectores: v1481389886, v1481389883, v1481390003, v1481390000, v1481389997, v1481389850, v2363396, v2363395, v2363394, v2363382, v1865629154, v1865630454, v1865630434, v1865630468. Unidad: LFS: persons in thousands, annual averages; Census: persons, randomly rounded to a multiple of 5; CSBC: percent of businesses.