A fallback procedure is a predefined backup process that takes over when a primary system fails, becomes unreliable, or is judged unsafe to continue. In cybersecurity, it is part of resilience planning: instead of stopping an essential service outright, the organization shifts to a safer manual mode, a redundant system, or a limited service path.
Fallback procedures matter because attacks often target availability and trust, not just data theft. If a model, application, or control system is compromised, defenders may need to disable automation, roll back to a known-good configuration, or require human approval until conditions are safe again. Good fallback design includes clear triggers, tested recovery steps, and logging so operators can tell whether they are recovering from a fault, a misconfiguration, or an intrusion. Without a fallback procedure, a single failure can cascade into a wider operational outage.