Evolving Test Processes in AI-Enabled Test Organizations CT-GENAI Practice Questions (Page 2)

Questions 6–10

1. 6application · medium

   A test team uses a generative AI tool to automatically generate test cases from requirements. During a review, a tester finds that the AI has produced a test case that would cause a destructive action in the production environment if executed. What is the most appropriate immediate action to mitigate this AI-specific risk?

   • ARun the AI-generated test case in a sandbox environment to observe its behavior.
   • BAdd a mandatory human review step for all AI-generated test cases before they are added to the execution suite.
   • CAdd a rule to the AI prompt to avoid generating destructive test cases.
   • DDisable the AI tool and revert to manual test case generation.
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A team has integrated an AI-based test execution tool that dynamically selects and prioritizes test cases based on code changes. After a sprint, the team notices that some high-risk areas were not tested because the AI deprioritized them. The test lead wants to evolve the execution workflow to prevent this. What should the test lead do?

   • ARemove AI prioritization and run the full regression suite every time to ensure complete coverage.
   • BIncrease the number of test cases the AI can select from by adding more tests to the suite.
   • CDefine a minimum risk threshold that forces inclusion of tests for areas above the threshold, regardless of AI prioritization.
   • DRetrain the AI model with more historical data to improve its risk assessment accuracy.
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A team uses generative AI to create test data for a new feature. The AI generates realistic-looking but synthetic customer records. The team is concerned that the synthetic data may not cover edge cases that exist in production. What is the most effective way to manage this test data risk?

   • ACompare the distribution of synthetic data against production data profiles and augment with targeted edge-case data.
   • BUse the synthetic data as-is and rely on the AI's training to have captured all relevant patterns.
   • CReplace all synthetic data with a small sample of production data to ensure realism.
   • DGenerate a larger volume of synthetic data to increase the chance of covering edge cases.
   Reveal Solution Discussion Select an answer first
5. 9foundation · easy

   How should test reporting change to reflect AI-enabled testing outcomes?

   • ABy removing all metrics because AI makes them irrelevant
   • BBy reporting only the time saved by using AI
   • CBy reporting only the number of defects found
   • DBy including metrics on AI model performance and confidence
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A test team has adopted an AI tool that predicts which test cases are most likely to fail based on code changes. The tool requires access to historical test results and code commit data. The team is concerned about the quality of the predictions because the historical data contains many gaps. What is the most appropriate way to adapt the test process?

   • AUse the AI predictions as-is and accept that some tests may be missed.
   • BSwitch to a different AI tool that does not require historical data.
   • CImprove data collection by ensuring all test executions and code changes are logged consistently before relying on AI predictions.
   • DRun the full regression suite in addition to the AI-selected tests to ensure coverage.
   Reveal Solution Discussion Select an answer first