Canada’s AI strategy promises up to 250,000 new jobs by 2031, but a separate rapid-adoption scenario projects years of stagnant employment before a possible recovery. The contrast matters: a long-term jobs target does not tell workers who might be left waiting in the meantime.
Watch Desk analysis
What happened
The Walrus examines the federal strategy, which commits more than C$2.3 billion to AI adoption, domestic capacity and public trust. AI Minister Evan Solomon said the 250,000-job figure comes from modelling based on OECD data and assumes attainable rates of AI adoption.
A separate Signal49 model, produced with the Future Skills Centre, looks at a scenario in which AI, robotics and automation are rapidly deployed across industries. It projects employment falling by 142,000 by 2032 and remaining stagnant for seven years. By 2032, it estimates 1.2 million fewer jobs than in its baseline scenario, with unemployment peaking at 10.3 per cent in 2033.
Those figures are scenario modelling, not a forecast of what will happen. The model assumes employment later recovers, with close to 3.9 million jobs added between 2032 and 2045, partly through government spending funded by economic growth. Its authors acknowledge that intervention is a major assumption. The article also notes that Statistics Canada researchers have found no broad evidence so far that generative AI is causing job losses across Canada.
Why it matters
The model’s warning is not simply that AI might destroy jobs. It describes a possible mismatch in timing: productivity and economic growth rise while hiring stalls, with gains and losses unevenly distributed across sectors. In that scenario, manufacturing jobs do not recover, while health care accounts for nearly 1.1 million of the later jobs gained.
That makes the strategy’s headline job-creation number only one part of the picture. The modelling does not establish that this outcome is inevitable, but it points to a practical question for policy: what support reaches workers if disruption arrives before new jobs do?
Our read
A jobs target is a promise about where the benefits might lead. It is not a transition plan. The model is too conditional to treat as a prediction, but too specific to wave away: its recovery depends on governments choosing to spend the proceeds of growth on services and infrastructure.
Canada’s strategy should be judged not only on whether it encourages AI adoption, but on whether it explains how workers and regions would be supported if the gains arrive unevenly. That is rather more useful than assuming the labour market will sort itself out, which is a plan chiefly notable for being written by nobody.
What to watch
- Whether the federal strategy adds concrete support for workers facing AI-related disruption.
- Whether the 250,000-job target is accompanied by measures of job losses, hiring and sector-level change.
- Whether the rapid-adoption scenario’s assumed government intervention appears in actual policy and spending.
Discussion spark: Should Canada’s AI strategy be judged against its 250,000-job creation promise, or should it also have to set out what support workers get if adoption stalls hiring before those jobs appear?
Sources and evidence
- Is Canada’s AI Job Plan Too Sunny? (13 November 2026, 11:30 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.