AI Minister Evan Solomon arrives at Tuesday's federal cabinet meeting in Ottawa.Justin Tang/The Canadian Press
Viet Vu is manager of economic research at the Dais at Toronto Metropolitan University. André Côté is executive director at the Dais.
Nine years ago, the Government of Canada launched the initial Pan-Canadian AI Strategy, an effort that made investments to cement Canada’s lead in producing leading AI research, later expanding the strategy to promote commercialization of AI-enabled products. Ottawa was largely successful in the first goal, but the country lagged the world translating that research success into economic prosperity.
This week’s planned announcement on a refreshed artificial-intelligence strategy comes three years and seven months after OpenAI unleashed ChatGPT. Most notably, the announcement also comes six months after the federal government’s self-imposed deadline.
Multiple media sources outline a strategy with wide-ranging objectives and activities. Called “AI for All,” it aims to speak to all Canadians about the benefits and opportunities of AI.
The question is, are Canadians on board with that positive vision of AI?
Opinion: Canada’s AI strategy is set to fail before it launches
The authors of the 2017 Pan-Canadian AI Strategy were working in a bygone era, sorting through hypothetical scenarios involving technology and applications that did not yet exist. Today’s authors face a slate of different, intersecting challenges: positioning Canada to keep pace with the “hegemons and hyperscalers” leading the development of AI, readying the economy to maximize technological gains and addressing public anxiety about the technology’s real (and imagined) effects.
With this in mind, the government is proposing a vision that supports widespread AI adoption, invests in AI innovators to build national champions, supports AI literacy and upskilling, builds out cloud computing and focuses on sovereignty and international alliances. These offer a promising foundation.
Yet it is not yet clear whether all Canadians want AI in their work, or their lives. This is something that Canada’s AI Minister Evan Solomon has acknowledged, remarking that, “[AI] adoption moves at the speed of trust.” Public opinion surveys find that Canadians broadly distrust AI. Local backlash against the new data centres needed to power AI are just one example. Consequently, the chances that Canadians will blindly buy the vision painted in this strategy are uncertain at best.
The strategy does see this issue of trust, and proposes to mitigate it by ensuring that Canadians aren’t harmed by misused AI. Those actions include updating privacy and consumer protection legislation and introducing a more streamlined version of a bill to protect Canadians in online spaces (focusing on children). It also promises to look for ways to ensure that AI-generated content is labelled as such whenever Canadians interact with it online. These are good steps, but for Canadians to embrace an AI for All vision, the government also needs to directly confront the fears that many hold.
Opinion: Canada is not likely to become a conventional AI power. But here’s how it can still win
The most significant is around the future of work. From Dario Amodei, co-founder of Anthropic, speculating openly about the “jobs apocalypse” to university graduates booing AI leaders in commencement speeches, anxiety about the impact of AI on jobs and livelihoods is palpable. While the evidence around labour-market effects to date is limited, Canadians must be able to trust that their elected government is tracking the issue and prepared to respond. The revamped strategy will reportedly include job investments alongside access to AI training, but it primarily addresses short-term disruptions for youth. It does nothing to assuage the deeper feeling that AI will cause intractable changes to the economy.
To start, Canadians’ concerns about AI need to be met by the government with hard evidence, not hype. Expanding the investments the government has made in Statistics Canada’s Artificial Intelligence and Technology Measurement Program, which tracks AI’s diffusion and impact in Canada’s society and economy, is a strong step. But beyond collecting data, the government must support a broader ecosystem of public conversations that make use of these data to inform the broader public so everyday Canadians can have their questions directly answered.
The government must also demonstrate that they’re taking a sober look at the long-term suitability of existing labour-market policies. If an AI hurricane is coming, will the Employment Insurance system, work force support programs and income support offer shelters for workers caught in the storm? These are questions policy-makers have to address.
Canadians must choose to embrace AI for the strategy to succeed. The strategy shows that the federal government broadly understands this challenge, but the critical task will be to bring Canadians along on the journey – not as helpless passengers, but as fellow sailors navigating this tumultuous and unknown water.