Agentic AI adoption is outpacing governance: report

Governance and workforce stakes for HR

Data readiness underlies much of the gap: 73% of Canadian organizations struggle with incomplete data despite 65% believing themselves data-ready, SAP reported — often the same employee records and workforce analytics agentic systems act on.

Uneven AI literacy and weakened critical judgment rank as the top two workforce risks cited worldwide. More than half (56%) of Canadian organizations say staff use unapproved AI tools at least occasionally, tied globally to data leakage or IP exposure.

To close that gap, SAP executives say governance must be built into how AI agents operate, not bolted on afterward. 

“This ensures that humans set the strategy while AI handles the orchestration and execution, all within a governed, auditable environment with humans in the loop at key moments,” said Anthony Robinson, Chief Technology Officer, SAP Canada. Sean Kask, Chief AI Strategy Officer at SAP, said organizations “face two challenges at once: risk that moves faster than most governance frameworks can keep up with, and value that is harder to measure than expected.”

Canadian business leaders had better step up if they hope to reap the benefits of agentic AI. That’s because 21% of HR professionals report that AI agents are already being used today, and 78% expect their organisation to use at least one AI agent by the end of 2026, according to a report from People Managing People.