OSFI Releases Bulletin on Managing Risks Associated with Generative and Agentic AI
August 06, 2026
The Office of the Superintendent of Financial Institutions (“OSFI”) recently published a bulletin on the use of generative and agentic artificial intelligence (“AI”).
OSFI noted in its bulletin titled ‘Generative and Agentic Artificial Intelligence: Implications for Technology, Cyber Security, and Operational Resilience’, that while AI can improve efficiency, automate processes, and support decision-making, it also introduces technology, cybersecurity, operational, and third-party risks that require effective oversight. As such, OSFI emphasizes the need for strong governance, clear accountability, and ongoing human supervision, and the integration of AI-related risks into enterprise risk management frameworks.
The bulletin also highlights the importance of controls surrounding software development, access management, cybersecurity, and third-party relationships. Organizations should monitor AI usage, protect sensitive information through appropriate controls, ensure oversight over AI-related third-party dependencies, and maintain contingency plans to support operational resilience.
While the bulletin does not directly apply to federally regulated pension plans, it serves as a reminder for plan administrators that AI innovation should be supported by strong governance, effective risk management, and robust cybersecurity practices.
Click on ‘More information’ below to read the bulletin:
.png)