A customer proposes deploying AI-suggested optimizations straight to production once they pass a local build. What symptom would most directly suggest this practice is causing problems?
Select an answer to reveal the explanation.
Short Explanation
If you push untested optimizations straight to production, the tell is production incidents that trace back to those changes. You lose the safety net that catches behavior changes before customers see them.
Full Explanation
The correct answer is the option describing production incidents traced back to optimization changes that were never regression-tested or reviewed before release. Deploying optimizations directly to production without regression tests or review removes the safety net that would normally catch a behavioural change before customers see it, so incidents tracing back to those changes are the direct symptom. The option about the build pipeline taking longer to complete each release is incorrect because a longer build time is unrelated to whether changes were reviewed or tested; it is a performance characteristic of the pipeline, not a safety issue. The option about developers spending more time reading AI-generated code explanations is incorrect because time spent reading explanations is normal review activity, not a symptom of skipping review. The option about an increase in the number of optimization candidates the tool identifies is incorrect because more identified candidates reflects the volume of analysis, not whether the changes that get applied are validated. For the exam, treat deploy-directly-to-production without review as an unsafe practice, and expect its symptom to show up as production incidents.