CyberCertPrepAI Security FundamentalsSecure AI Deployment & MLOpsEASYFree question
A startup deploys their LLM application without any logging. When a user reports a potential policy violation, the team cannot investigate. Beyond compliance failure, what specific security operational capability is lost without logging?
- A.B. The ability to detect and respond to attacks (injection attempts, jailbreaks, extraction attacks) - without logs there is no detection capability, making the system effectively blind to ongoing exploitation
- B.A. The ability to improve model accuracy over time
- C.C. The ability to A/B test different model versions
- D.D. The ability to measure the model's average response latency
Why A is correct
Logging is a foundational security capability (NIST CSF: Detect, Respond). Without logs: injection attacks can continue indefinitely, jailbreaks can be systematically exploited without detection, extraction attacks can complete before anyone notices, and incident investigation becomes impossible. This maps to OWASP LLM07: Inadequate AI Logging and Monitoring (in later versions of the OWASP LLM Top 10). Model accuracy improvement, A/B testing, and latency measurement are operational concerns, not security capabilities - though logs support all of them.
Know someone studying for AI Security Fundamentals? Send them this one.