The Carney government's AI strategy is a bet that adoption can outrun public anxiety. But the anxiety is broad, deep, and in some places already organized, and the political battlelines are about to be drawn across jobs, privacy, misinformation, and the data centres going up in people's backyards. Companies adopting AI need to be thinking about the politics of how they do it, not just the economics.
Key Takeaways:
- Canada's "AI for All" strategy is a bet that adoption can outrun public anxiety. It spends heavily to lift business uptake from 12 to 60 percent by 2034 whilst leaving the privacy, safety, and labour protections meant to earn public trust for later. That sequencing, adoption now and guardrails eventually, creates political risk for both the government and corporate AI adopters.
- Canadian wariness of AI is broad, deep, and crosses both class and party lines, and on the most visible front, the data centres going up in Hamilton, Vancouver, and elsewhere, the backlash has already arrived. What it has not yet found is a national political champion. When it does, the firms that adopted AI most conspicuously at labour's expense will be targets.
- Companies should treat the politics and optics of AI as seriously as the economics, and across every front rather than layoffs alone: how they handle data, whether their automated decisions are explainable, the creative workers whose output trained the models, and where their compute gets built. The framework building phase has started in the public sphere, notably Pope Leo XIV's first encyclical, the Privacy Commissioner's ruling against ChatGPT, and data-centre fights spreading across the country, and the firms that anticipate it will fare better than the ones it catches.
