Making Sense.

Two separate procurement signals from Canadian governments arrived within a week of each other this month. Toronto-based Cohere signed a memorandum of understanding with Québec, its first agreement of its kind with a provincial government. That arrangement is exploratory and carries no contract or funding yet, focusing instead on planning for digital sovereignty and responsible AI across the public service. Days earlier, the Canadian Centre for Cyber Security gained access to Anthropic's Mythos model.

These were two separate choices by two different levels of government, not one coordinated strategy. These agreements hint at where procurement expectations are heading. Buyers are placing more weight on data residency, on sovereignty, and on the ability to match a particular tool to a particular task.

A new survey from Express Employment Professionals and The Harris Poll captures a widening divide between how Canadian employers and their staff view AI at work. Among the hiring managers surveyed, 63% report that their company already uses AI in some form, and most expect it to improve efficiency and free up their teams' time. The workers themselves turned out to be considerably more wary. Among workers, 94% reported concerns about AI in the workplace, 78% worried their employer would hire fewer people because of it, and 75% of those already at companies using AI feared eventual workforce cuts.

When a team suspects that AI is meant to replace them, leaders tend to get quiet resistance rather than the productivity gains they expected. The people closest to the work also stop pointing out where the tools genuinely help and where they fall short. The organizations earning real trust treat adoption as a change in how people work, and they frame AI as a multiplier of what their people can do. Since the tools are changing so quickly, they lead with durable principles instead of fixed rules about what AI can and cannot do, and they put real money into training rather than assuming staff will work it out on their own.

This month Clio, a Burnaby-based, AI-powered legal practice management software company, acquired Toronto's Jurisage for its AI-ready legal datasets in Canada, as Clio moves its AI from helping run a practice toward handling the legal work itself.

For leaders across professional services, the deal shows how AI competition in these fields increasingly turns on proprietary data and the speed of locking it up. Firms in law, accounting, and advisory work now face a question that runs deeper than whether their tools happen to include AI. The more important question is whose data sits behind those tools, and what that dependence means for their position over the next few years.