Ahead of a production rollout, why is it risky to apply every optimization suggestion a code analysis tool produces without testing?
Select an answer to reveal the explanation.
Short Explanation
Faster isn't the same as correct. You should test every optimization before it goes anywhere near a rollout, because a suggestion can change behavior in ways only testing reveals.
Full Explanation
The correct answer is the option saying some suggestions may change the program's behaviour in ways only testing would reveal, so applying them untested risks shipping a functional regression. An optimization suggestion is a hypothesis about improving code, and only testing confirms whether applying it changed behaviour along with performance. The option saying tools generate only a small number of suggestions per run is incorrect because suggestion volume is not the reason untested application is risky. The option saying suggestions cannot be reviewed by more than one engineer at a time is incorrect because review capacity is a process detail, not the underlying risk. The option saying applying suggestions always increases codebase size is incorrect because optimizations frequently reduce code size or complexity rather than increase it, and size is not the concern here regardless. For the exam, the risk of untested optimization is behavioural change, not anything about tooling limits.