AI Oversight Lags Behind Deployment in Canadian Organizations
A new study from IBM finds that Canadian organizations are deploying AI faster than they can govern it. More than 1,000 senior leaders across 20 countries, including Canada, were surveyed about how their organizations manage AI as it moves from testing into critical operations.
The findings arrive as Canadian regulators, boards, and the public increasingly scrutinize how AI influences decisions in healthcare, transportation, financial services, and public programs.
Governance Gaps Are Costing Money and Control
Sixty-three percent of Canadian executives say gaps in AI governance already slow their ability to deploy AI at scale.
Large Canadian enterprises lose an estimated $144 million per year to AI errors, bias, duplication, and uncoordinated deployments. Half of those losses stem from governance gaps-not from the technology failing, but from organizations lacking visibility into which AI systems they use and how those systems behave.
"You can't govern systems you can't see," said Manav Gupta, Vice President and CTO of IBM Canada. "AI systems now act as critical infrastructure, and that raises real questions about trust, accountability, and sovereignty."
Control Matters More Than Data Location
Discussions of digital sovereignty in Canada often focus on where data lives or who owns it. The study points to a more immediate operational challenge: organizations need to see which AI tools they use, understand how decisions get made, control access and updates, and intervene when systems behave unexpectedly.
As AI influences decisions that affect people and services, organizations face growing pressure to explain who controls these systems and who is accountable when outcomes go wrong.
"Digital sovereignty is about control, not isolation," Gupta said. "Organizations need governance they can enforce and demonstrate, not just policies on paper."
Few Organizations Coordinate AI Governance Across Operations
Organizations that coordinate AI governance across the full lifecycle of AI systems-from deployment through monitoring-report stronger productivity gains, higher returns on investment, and fewer losses tied to AI errors.
Only 18 percent of Canadian organizations say they currently have systems in place to coordinate and govern AI across everyday operations.
The gap between policy and operational reality remains wide. Many technology environments struggle to provide consistent, auditable answers about how AI systems work and who controls them-exactly what regulators, auditors, and boards now expect.
For operations professionals, this means governance is no longer optional. AI for Operations requires clear visibility into systems, documented control procedures, and the ability to act when something goes wrong. AI Learning Path for Operations Managers can help teams understand how to build that control into their operations.
The full IBM study, AI in motion: Orchestrating AI at scale for sovereignty and resilience, is available on IBM's website.
Your membership also unlocks:
