Carney's AI Council Wants Canada to Build Its Own AI Future

Ottawa is assembling a brain trust to steer Canada's AI ambitions. Prime Minister Mark Carney on October 2, 2026 announced the Prime Minister's National Council on Artificial Intelligence, a 13-member advisory panel meant to guide the federal AI for All strategy through its next phase. The move puts some of the country's best-known AI thinkers inside the policy tent as Canada tries to turn research strength into real economic power.

The new AI council will bring together public and private stakeholders and advise on how the government's national strategy evolves, according to the official announcement. Its mandate covers accelerating adoption, fostering homegrown champions, and building sovereign AI infrastructure. Members will also weigh in on making AI safer, protecting democratic processes, and ensuring the technology's benefits reach all Canadians. A small team inside the Privy Council Office will support the panel, while ministers remain accountable for actual policy decisions.

The roster reads like a who's who of Canadian AI. It includes deep-learning pioneer Yoshua Bengio, innovation economist Ajay Agrawal, AI safety researcher Gillian Hadfield, Vector Institute co-founder Sanja Fidler, and Cloudflare co-founder Michelle Zatlyn. Former Bank of Canada governor's deputy Louis Vachon, Senator Thomas Pitfield, and former national security adviser Jody Thomas are also on the list, alongside tech investor Patrick Pichette in an ex officio role. Carney said the panel exists because prosperity and sovereignty in the age of AI belong to nations that can build, adopt and govern the technology on their own terms.

What AI for All is actually trying to do

The council arrives four months after Carney unveiled AI for All in June 2026, and the two moves only make sense together. The strategy commits to generating an additional two hundred billion dollars of economic growth and creating a quarter million AI-related jobs over five years, according to reporting on the federal plan. It also sets a target of lifting business AI adoption from just over 12 percent to 60 percent by 2034. For young Canadians specifically, the government promises up to 90,000 AI-related jobs and work placements, plus training meant to reach one million entry-level post-secondary students.

The timeline tells the story of a government sprinting. June brought the strategy launch at the University Health Network in Toronto. September saw the debut of Digital Transformation Canada, a new federal body charged with modernizing government services and using public purchasing power to help domestic digital and AI firms scale. Now October's AI council supplies the outside expertise. The sequencing is deliberate: strategy first, delivery machinery second, independent advice third.

Sovereignty is the watchword

Sovereignty is the word Carney keeps returning to, and the council's job description makes it official. One of the strategy's three pillars is strengthening Canada's control over critical AI capabilities, alongside building trust and expanding opportunity. That language matters because Canada, despite its AI research pedigree, still leans heavily on foreign models, cloud platforms and chips. The government is pushing to build sovereign Canadian foundations in compute, cloud, connectivity, data and talent, including a public AI supercomputer.

Corporate Canada is already moving on that signal. Bell and Cisco recently signed a deal to host AI capacity in Canada, part of what analysts describe as a broader sovereign AI push with real workforce stakes. Toronto-based Cohere has pitched itself as a trusted domestic choice for enterprise buyers, and it announced a planned merger with a German AI developer earlier this year. For young jobseekers, the bet is that this infrastructure creates an actual job market at home rather than a brain drain pipeline south.

The trust problem this panel has to solve

Here is the catch: Canadians are not sold on AI yet. The Ipsos AI Monitor 2026 found that two-thirds of Canadians say AI makes them nervous, and fewer than one in five are comfortable with AI screening job applicants. Only 46 percent of Canadian workers said AI saved them time in the past year, compared with 62 percent globally. That trust gap is precisely why the strategy names trust as its first pillar, and it is the audience this council is really performing for.

Skeptics will also note what the panel cannot do. It advises; it does not decide. Policy calls and implementation stay with ministers and departments, which means the council's influence depends entirely on whether the government actually listens. And the federal approach is deliberately light on rules: the government has moved away from reviving binding AI legislation, leaning instead on education, standards, sectoral oversight and targeted measures on deepfakes and labelling. Whether persuasion beats regulation in keeping AI accountable is an open question worth watching.

Still, for a generation that will live inside AI-transformed workplaces, the next test is concrete: watch how many of those promised placements materialize, whether adoption climbs toward the 2034 target, and how quickly sovereign computing capacity comes online. The council has been named. Now the deliverables have to start landing, and this AI council will be judged on delivery, not announcements. Catch up on more Canadian politics in our Canadian News hub, including how housing is shaping Toronto's mayoral race.