Who Answered
Canada's AI Strategy Consultation
Over 11,300 people took part. They answered 26 questions in October 2025, and a classification pipeline built on four large language models sorted more than 64,600 answers into themes.
Use the arrow keys or the buttons below.
ISED's largest public consultation, by its own count
Three channels fed the strategy: an online survey open from 2025-10-01 to 2025-10-31, written reports from the AI Strategy Task Force, and unsolicited policy submissions.
Source: ISED, Engagements on Canada's Next AI Strategy: Summary of Inputs, Q1 2026, Cat. Iu4-453/2026E-PDF
Most respondents answered as individuals
Respondents classified themselves twice: once by participation type, once by role. The report gives percentages for both but does not state how many people answered either question.
View as table
| Participation type | Share |
|---|---|
| As an individual | 83% |
| For an organization | about 17% |
View as table
| Role | Share |
|---|---|
| Interested Canadian | 52% |
| Canadian business representative | 19% |
| Academic or researcher | 13% |
| Other | almost 6% |
| Government or regulator | 4% |
| Industry association | 3% |
| Privacy advocate or legal professional | 2% |
British Columbia's share was 1.5 times its population. Quebec's was about a third.
Ontario's share of responses matched its share of Canada's population. The other provinces split above and below. Nunavut does not appear in the published list.
View as table
| Province or territory | Response share (%) | Population share (%) | Ratio (Northern Signal calculation) |
|---|---|---|---|
| Ontario | 39 | 38.9 | 1.00 |
| Quebec | 7.6 | 21.8 | 0.35 |
| British Columbia | 20.6 | 13.7 | 1.51 |
| Alberta | 7.8 | 12.1 | 0.64 |
| Manitoba | 1.9 | 3.6 | 0.52 |
| Saskatchewan | 1.4 | 3.0 | 0.46 |
| Nova Scotia | 3.8 | 2.6 | 1.45 |
| New Brunswick | 1.45 | 2.1 | 0.69 |
| Newfoundland and Labrador | 0.9 | 1.3 | 0.68 |
| Prince Edward Island | 0.26 | 0.44 | 0.59 |
| Yukon | 0.16 | 0.12 | 1.38 |
| Northwest Territories | 0.06 | 0.11 | 0.54 |
| Nunavut | 0.10 | Not listed in source |
IT, tech and cyber was the most-named sector
1,860 participants named a sector for their organization. Arts, entertainment and recreation came third, ahead of academia and health care.
View as table
| Sector | Selections | Share of 1,860 (%) |
|---|---|---|
| IT, tech and cyber | 645 | 35 |
| Professional, scientific and technical services | 372 | 20 |
| Arts, entertainment and recreation | 272 | 15 |
| Academia | 235 | 13 |
| SME | 223 | 12 |
| Health care | 215 | 12 |
| Information and cultural industries | 193 | 10 |
| Government | 174 | 9 |
| Financial sector | 113 | 6 |
| Manufacturing | 77 | 4 |
| Natural resources | 60 | 3 |
| Administrative | 42 | 2 |
| Construction | 39 | 2 |
| Retail trade | 36 | 2 |
| Transportation and warehousing | 29 | 2 |
| Hospitality | 26 | 1 |
| Real estate | 22 | 1 |
Half of those who gave an age were 35 to 54
3,146 participants gave an age group. 1,599 of them, or 50.8%, were between 35 and 54. Four were under 18.
View as table
| Age group | Respondents | Share (as published) |
|---|---|---|
| Under 18 | 4 | 0% |
| 18–24 | 116 | 4% |
| 25–34 | 665 | 21% |
| 35–44 | 931 | 30% |
| 45–54 | 668 | 21% |
| 55–64 | 361 | 11% |
| 65+ | 156 | 5% |
| Prefer not to say | 245 | 8% |
Each profile table describes a minority of participants
The location, age and sector tables each rest on a different subset. None covers more than about 28% of the 11,300-plus participants, and the report does not say whether the subsets overlap.
View as table
| Breakdown | Respondents | Share of 11,300 (upper bound) |
|---|---|---|
| All online participants | over 11,300 | |
| Location (organization) | 3,132 | 27.7% |
| Age group | 3,146 | 27.8% |
| Sector (organization) | 1,860 | 16.5% |
A model pipeline, with human review, assigned 64,600 answers to themes
The report describes two tools and a review step. It also states that pipeline outputs were "paraphrased or taken directly" into the report's own text (p. 16).
Commercial survey analytics
SimpleSurvey's natural language processing for sentiment, keywords and topics.
Internal classification pipeline
Built by ISED. Cleans responses and maps them to a taxonomy of themes and subthemes.
Human review
"At several stages." Reviewers validated and refined the generated analyses.
Report drafting
Pipeline summaries used in the published text. The same method applied to the Task Force reports.
The report's only accuracy figure. It states the solution had "at least a 90% success rate" in categorizing responses into specific intents (p. 16). The document does not define the figure. It does not say:
- what counted as a success
- how many responses were checked, or how they were chosen
- whether the rate applies to one model, the ensemble, or the reviewed output
- how disagreements between models were resolved
Source: ISED, Summary of Inputs, Q1 2026, pp. 5, 16. Northern Signal has not estimated the rate.
The tables now sit beneath a new advisory council
The Prime Minister's National Council on Artificial Intelligence will advise on how the AI for All strategy evolves. This consultation is part of the record that strategy was built on.
In a nutshell
The short answers, with the same figures and the same caveats as the slides.
How many people took part in Canada's AI strategy consultation?
Over 11,300 people sent more than 64,600 responses to 26 questions across 8 pillars in ISED's online survey, open from 2025-10-01 to 2025-10-31. The strategy also drew on 32 AI Strategy Task Force reports and close to 300 supplementary submissions.
Who answered the consultation?
By participation type, 83% answered as individuals and about 17% on behalf of an organization. By role: interested Canadians 52%, business representatives 19%, academics or researchers 13%, government or regulators 4%, industry associations 3%, privacy advocates or legal professionals 2%, and other almost 6%. The report does not say how many people answered either question.
Which provinces were over- or under-represented?
Based on organization location (n = 3,132), British Columbia's response share was 20.6% against 13.7% of Canada's population, a ratio of 1.5. Ontario was proportional at 39% against 38.9%. Quebec was 7.6% against 21.8%, about a third of its population share. The ratios are a Northern Signal calculation, and about 13% of location responses are not accounted for in the published list.
Which sectors responded most?
Of 1,860 participants who named a sector, 645 (35%) chose IT, tech and cyber, 372 (20%) professional, scientific and technical services, and 272 (15%) arts, entertainment and recreation. Counts sum to 2,773, which indicates respondents could pick more than one sector.
How old were the respondents?
Of 3,146 participants who gave an age group, 1,599 (50.8%) were between 35 and 54, 245 preferred not to say, and 4 were under 18. The 50.8% share is a Northern Signal calculation from the published counts.
Do the published tables describe everyone who took part?
No. The location table covers 3,132 participants, the age table 3,146 and the sector table 1,860. None covers more than about 28% of the 11,300-plus participants, and the report does not say whether the subsets overlap.
How were 64,600 answers sorted into themes?
ISED used SimpleSurvey's natural language processing plus an internal classification pipeline built on Cohere Command A, OpenAI GPT-5 nano, Anthropic Claude Haiku and Google Gemini Flash, with human review at several stages. The report states "at least a 90% success rate" in categorizing responses into specific intents but does not define the figure.
Where every number comes from
- Engagements on Canada's Next AI Strategy: Summary of Inputs. Innovation, Science and Economic Development Canada, Q1 2026. Cat. Iu4-453/2026E-PDF, ISBN 978-0-660-97978-6, 19 pp.Primary source for slides 2 to 8. Opened 2026-10-14. Landing page: ised-isde.canada.ca, opened 2026-10-14.
- Statistics Canada, Table 17-10-0009-01, Population estimates, quarterly. Preliminary estimates for 2025-10-01, transcribed from Nunavut Bureau of Statistics StatsUpdates for Q3 and Q4 2025.Verify against tablePopulation shares on slide 4. Opened 2026-10-15.
- Government of Canada launches AI Strategy Task Force and public engagement. ISED, September 2025.Consultation launch, slide 9. Opened 2026-10-14.
- Prime Minister Carney launches AI for All. PMO, 2026-06-04.Strategy launch and February 2026 release date. Opened 2026-10-15.
- Prime Minister Carney launches new National Council on Artificial Intelligence. PMO, 2026-10-02.Council launch and membership, slide 9. Opened 2026-10-14.
Method
Figures are reproduced as published, including rounding. Northern Signal calculations are limited to the ratio view on slide 4, the 35-to-54 share on slide 6 and the shares of total on slide 7, and are labelled where they appear. No figure in this deck is estimated. Northern Signal takes no position on the consultation's findings.
Read with: Canada's AI for All Strategy, Explained, the strategy this consultation fed.
Licence
Source tables: © His Majesty the King in Right of Canada, 2025. Reproduced under the publication's Permission to Reproduce, with the Department of Industry identified as source institution. This is not an official version and was not made in affiliation with, or with the endorsement of, the Department. Commercial reproduction requires Crown copyright clearance.
Northern Signal compilation, calculations and charts: CC BY 4.0.
Northern Signal tracks Canadian AI policy, funding and adoption every week, in one free email. New Atlas dossiers go to readers first.
Get the weekly briefing