Loading...
Loading...

When most Canadian financial institutions talk about Guideline B-13, the conversation gravitates toward cybersecurity: governance, threat detection, incident response, controls against ransomware, and credential theft. That's the visible half of the technology risk picture. The other half - operational software defects - gets far less airtime in B-13 discussions, even though those defects are responsible for a substantial share of real-world service disruptions in the financial sector.
This is a gap worth closing, as OSFI's expectations in this area are more explicit than many institutions realize.
Guideline B-13, in force for FRFIs since January 1, 2024, sets out OSFI's expectations for how institutions identify, assess, manage, and remediate technology and cyber risks in support of operational resilience. The guideline is principles-based and organized around three domains, with the Technology Operations and Resilience domain carrying the bulk of the operational-defect-relevant expectations.
Within that framework, institutions are expected to maintain sound practices across four disciplines that bear directly on non-security software defects:
Change management: controlling what gets deployed, when, and under what conditions, including vendor-supplied updates
Technology asset management: maintaining accurate inventories of what's running where, and at what versions
Incident management: detecting, responding to, and learning from operational disruptions regardless of cause
Vulnerability management: identifying defects in the technology stack that could cause harm
It's the last item that often gets misread. In many institutions, "vulnerability management" has come to mean "CVE patching" - narrowly focused on security vulnerabilities tracked through standard cybersecurity feeds. B-13's language is broader than that, and OSFI has confirmed as much in direct correspondence: institutions are expected to have appropriate controls and processes to manage the risks arising from the technology they use, including software issues that could lead to operational impacts.
A non-security software defect doesn't necessarily look like a cyber threat. There's no attacker, no exploit, no CVE identifier, no urgency from the security operations center. What there is, instead, is a defect buried in a vendor's release notes or support bulletin that can corrupt a database, crash a payment processing service, or degrade performance below tolerance for hours at a time.



Kevin Roche
April 3rd, 2026

Sign up to receive a monthly email with stories and guidance on getting proactive with vendor risk
BugZero requires your corporate email address to provide you with updates and insights about the BugZero solution, Operational Defect Database (ODD), and other IT Operational Resilience matters. As fellow IT people, we hate spam too. We prioritize the security of your personal information and will only reach out only once a month with pertinent and valuable content.
You may unsubscribe from these communications at anytime. For information on how to unsubscribe, as well as our privacy practices and commitment to protecting your privacy, check out our Privacy Policy.

Nigel Moulton
February 11th, 2026
The customer impact is identical to a cyber-driven outage. The regulatory exposure under B-13 is comparable, but the institutional response is often nowhere near as mature, because the data feeds, ownership models, and remediation workflows built for CVEs don't automatically extend to operational defects.
That asymmetry creates risk. A bank may have a best-in-class CVE management program, but without an equivalent program for vendor operational bugs, it is effectively managing only part of its software risk surface. The portion that remains unmanaged is often the very portion most likely to cause service disruptions, outages, and operational failures.
OSFI's position is not an outlier. Across jurisdictions, financial regulators are converging on the view that operational resilience requires managing third-party software risk in its entirety, not just its security subset.
European Union: DORA, fully enforceable since January 2025, requires financial entities to manage ICT risk across the full lifecycle of vendor software, regardless of whether the trigger is a security event or an operational defect.
United Kingdom: In a joint regulatory case study, the FCA confirmed that, under its SYSC 15A operational resilience rules, authorized financial services firms must manage risks in software and applications through an established risk framework, process and procedures, regardless of whether they are caused by security or non-security known or unknown bugs.
UK Digital Regulation Cooperation Forum: Four UK regulators (Ofcom, FCA, CMA, ICO) jointly published a case study on the operational resilience risks of non-security third-party software defects.
Canada: OSFI's B-13, as discussed above, expects institutions to manage software issues that could lead to operational impacts.
The boundary between "security" and "operational" software risk is dissolving in regulatory language because, in the real world, it never meaningfully existed for customers affected by an outage.
For Canadian banks, insurers, and other federally regulated institutions, three practical steps follow:
Audit your current scope. If your existing technology and cyber risk program treats security vulnerabilities and operational bugs as separate problems with different owners, or treats non-security bugs as a development concern rather than a risk management concern, you likely have a gap to close.
Map operational defects to your B-13 controls. Specifically: change management, technology asset management, incident management, and vulnerability management. Each of these disciplines should explicitly account for non-security vendor defects.
Build the evidence trail. B-13 is principles-based, but supervisors will still want to see how you're operationalizing those principles. Real-time dashboards, ITSM integration, and auditable remediation workflows turn a policy statement into a demonstrable program.
This is the work BugZero was built for. Our platform aggregates non-security operational defect data from across your vendor stack, maps each defect to the production devices, systems, and applications that it affects, and prioritizes remediation based on business impact. The result is exactly the kind of evidence-backed, risk-based, continuously-monitored program that Guideline B-13 is intended to drive institutions toward.
The regulatory direction is set. The question for Canadian FRFIs is how quickly the operational practice catches up.

Nigel Moulton
February 11th, 2026