Canada’s AI strategy to be released as survey suggests public trust gap

The federal government is set to release its highly anticipated artificial intelligence strategy Thursday morning after months of delays and hints at what it might contain.

The strategy’s release will come on the heels of a new Ipsos report released Wednesday, which found Canada remains among the least enthusiastic countries in the world about AI — and that Canadians’ excitement has dropped since last year.

Artificial Intelligence Minister Evan Solomon says the strategy will be titled “AI for All” and will focus on unleashing the economic potential of AI while building Canadians’ trust in the technology — which the Ipsos report suggests is also at a global low.

“That means AI is going to benefit every Canadian, no matter where you live, no matter what you do, no matter what age you are,” Solomon said in a social media video posted Wednesday.

“I know there’s a lot of anxiety and a lot of excitement about AI. Our plan is very practical, it’s very pragmatic. It’s not about being a cheerleader, it’s not about throwing pitchforks either. It’s about building a transparent, practical plan.”

A government source told Global News on background that the strategy will contain “a big push on adoption” of AI across Canadian private and public sectors.

“Broad pillars” of the plan will include ensuring sovereign control of AI technology and “harnessing the opportunity this new technology presents,” the source said.

The government’s spring economic statement in April outlined six “pillars” the AI strategy will be based on, with “protecting Canadians and safeguarding our democracy” at the top of the list.

Other pillars include providing access to AI literacy and skills training, accelerated AI adoption, supporting the building of “sovereign compute infrastructure at scale,” and international co-operation.

Solomon told reporters Tuesday the strategy will be followed by new legislation on online harms and updates to Canada’s private sector privacy laws. The government source said the strategy will also have details on measures to criminalize sexualized AI-generated deepfake images.

“We will have … comprehensive legislation that will make sure that we’re protecting children first and foremost … Canadians’ privacy and Canadians’ data,” Solomon said.

Solomon has said the strategy will also consider the technology’s impacts on the labour market.

NDP parliamentary leader Don Davies said Tuesday the potential for AI to create mass unemployment if left unchecked, as well as mounting safety concerns around tools like AI chatbots, requires the government to move more swiftly.

“I think our government is asleep at the switch, as it were, because I think that technology is moving much more rapidly and we don’t have any guardrails,” he told reporters.

Get daily National news

Get daily Canada news delivered to your inbox so you’ll never miss the day’s top stories.

“We can harness the best parts of AI and use that technology for positive purposes, but at the same time I think it’s … completely unregulated and terribly dangerous if we don’t get some guardrails up on it soon.”

Ipsos’s annual AI Monitor report, released Wednesday, surveyed 32 countries on attitudes around AI and found Canadians remain among the least-trusting and least-enthusiastic populations when it comes to the technology.

The report found the number of Canadians who expressed excitement about AI products and services actually fell five points since last year, to just 26 per cent.

Canada topped the list of countries whose citizens felt nervous about AI, with 67 per cent saying so — 17 points above the 32-country average.

Ipsos also found Canada had the lowest share of people among all countries surveyed who believe AI will improve their country’s economy (14 per cent) or the job market (11 per cent) in the next three to five years.

Only 27 per cent of Canadians said they trust companies that use AI will protect their personal data, according to the report.

AI industry and research leaders have expressed concern that, in his first year as AI minister, Solomon has largely prioritized the potential for AI adoption and growing the Canadian AI industry over the associated safety risks posed by the technology.

Other experts have since said Solomon and the government appear to have recognized the need to shift those priorities and recognize the gap between low public trust and private sector enthusiasm.

Shion Guha, an assistant professor at the University of Toronto who studies the intersection between AI and government policy, told Global News he’s “very cautiously optimistic” the strategy will find an appropriate balance.

“It’s trying to thread a needle with a very small eye hole,” he said. “It’s quite possible to do that.

“I think the way to do that is to figure out what governance frameworks we need to have around AI. Right now we don’t (have those).”

Seeing concrete regulations passed through legislation would go a long way toward building Canadians’ trust, Guha said, rather than “the government saying, ‘Please trust us.’”

“Just saying that we have a strategy and different pots of money and different things that different government agencies will do without figuring out those governance pathways will not build trust,” he added.

In a separate, forthcoming global study comparing AI attitudes among workers and employers, the AI-powered employment platform Employment Hero found just 41 per cent of Canadians feel confident their AI skills are sufficient to compete in the job market.

Employment Hero released details from the report to Global News ahead of its full release on Thursday.

The “AI Paradox Report” also found 20 per cent of Canadian business leaders say AI skills are essential when considering prospective employees, while 30 per cent are ranking it ahead of a post-secondary degree as a prerequisite.

Megan Felsing, the company’s head of communications, said in an interview that the AI strategy’s promise of widespread literacy and skills training will be critical to address the report’s findings.

She noted that, according to the study, 58 per cent of Canadian workers are gaining their AI skills independently through social media platforms.

“I’m really interested to see what these training plans actually look like in practice,” Felsing said, particularly if they are available for free.

She added that the strategy should also provide guidance to employers on how to encourage AI usage in the workplace so that “employees know that it’s not wrong and it’s not cheating to use AI.”

Richard St-Pierre, a senior advisor on quantum and digital sovereignty at Quebec business technology firm Levio, told Global News that the strategy must also balance AI sovereignty with AI’s economic potential.

He said pursuing sovereign AI products and infrastructure, while essential, shouldn’t come at the expense of harnessing AI to address Canada’s widening productivity gap with other G7 countries.

“AI is viewed as a technology thing, but at the end of the day it’s an economic race,” he said.

“If we say we’re going to wait to have a completely sovereign AI Canadian solution before we start that inflection point (of) upward movement, we will continue to lag behind.”

Prime Minister Mark Carney as well as Solomon have pointed to AI as a key way to improve Canada’s productivity metrics, something St-Pierre also agreed with.

“I cannot name another big economic activity — whether it’s manufacturing, airlines, whatever — that will have that breadth and depth of impact AI will have on productivity and potentially flip the productivity gap in the positive side,” he said.

“That’s where AI is at the forefront.”