Start a Project

AI

Canadians are turning against AI. How your business can still earn trust

Ottawa is spending billions on AI while public trust falls. For a Canadian business, trust is now the product. Here is how to earn it.

A small Canadian flag beside a stack of reports and a glass shield on a light wooden desk

Ottawa is spending billions on AI while Canadians grow more sceptical of it. A new poll shows that one in four Canadians now names AI as one of the greatest threats to the country, up from one in ten only six months ago. For a Canadian business that uses AI, that number is not background noise. Your customers are deciding right now whether they trust you with their data, their money and their time.

What the numbers say

Abacus Data found that 26 percent of Canadians now select AI as a threat, up from 10 percent in April 2026. AI is now the second most frequently selected threat in the survey, behind Trump and the United States. Among voters who strongly approve of the Carney government, 36 percent select AI. The poll was published in the days around Budget 2026, which was tabled in Ottawa on 5 November and puts AI at the centre of the government's growth plan, with 2.6 billion dollars in spending tied explicitly to AI related programs and a promise of access to 500 megawatts of sovereign AI compute for Canadian firms.

The government is also backing Canadian AI directly. On 29 October in Montreal, Prime Minister Mark Carney announced an equity investment of up to 1.6 billion Canadian dollars in Cohere, the Toronto founded AI company. Cohere agreed to keep its headquarters in Canada, create at least 1,000 new technical jobs and commit at least 1.7 billion dollars to computing infrastructure in Canada by 2030. On 4 November, Cohere said its multi year effort in Canada is now expected to create more than 1,500 jobs and deliver more than 4 billion dollars in research and infrastructure investment through 2031. Canada's AI minister, Evan Solomon, spent 3 November in Berlin at Germany's Digital Summit, talking about sovereign AI cooperation between the two countries.

Why trust is falling

The backlash has clear causes. Data centres are being built in communities that were not asked first. AI tools now read bank accounts and act on people's behalf, which feels close to home. And most people have never been told, in plain words, what a business does with their data. When the government promises 90,000 AI related opportunities by 2031 while a quarter of the public calls AI a threat, the gap is a communication failure as much as a policy question.

What earns trust

Trust is not won with a slogan. It is won with specific, checkable behaviour. Five things work.

  1. Show a human. Put a named person behind every AI answer your customers see, and make it easy to reach that person. A chatbot that hides the team behind it reads as a wall.
  2. Say what data you use, in one paragraph, on the page where the AI runs. No legal fog. If you use a customer's message to improve your service, say so, and say how they can opt out.
  3. Publish your AI rules. Write down which tools you use, what they may do and what they may never do, for example never move money and never share data with a third party. Put it where customers can find it.
  4. Let customers choose. Offer the human route next to the AI route, every time. People accept AI when they chose it. They resent it when it is forced on them.
  5. Fix mistakes in public. When your AI gets something wrong, say what happened, what you changed and who to contact. One honest correction builds more trust than ten perfect weeks.

A simple test before you launch anything with AI

Before any AI feature goes live, answer four questions in writing: what data does it touch, who can see the result, what happens when it is wrong, and how does a customer reach a human. If any answer is unclear, the feature is not ready. This test takes twenty minutes and prevents the kind of story that ends up in a poll about threats.

Canada is investing heavily in AI, and that will not stop. The businesses that win customers in this climate are the ones that treat trust as part of the product, not as a message written after the product is built. Be specific, be reachable, and let the customer decide.

See our AI chatbot service and our company branding service. For the budget context, read our Canada Budget 2026 guide.

Frequently asked questions

How many Canadians see AI as a threat?

The Abacus Data poll published around Budget 2026 found 26 percent of Canadians select AI as a threat, up from 10 percent in April 2026. It is the second most cited threat, behind Trump and the United States.

Does the backlash mean AI investment will stop?

No. The same budget that funds AI also funds the compute, training and jobs around it. The point is that public support cannot be assumed. Businesses that explain their AI clearly will find customers more willing than businesses that hide it.

What is the first step for a small business?

Start with one page: what data you use, what your tools may and may not do, and how to reach a human. Then apply the four question test in this article before every new AI feature.

Does using Canadian AI tools solve the trust problem?

It can help, because Canadian built AI keeps data and jobs in the country. But a Canadian label alone does not create trust. Customers still want to know what happens to their data and who is responsible when something goes wrong.

Working on something like this?

Tell us what you are building. We reply to every enquiry.

Related service: AI Search Visibility (GEO)

Start a Project