Detecting errors in AI output · Question 53 of 60
For a banking app, an LLM suggests only test cases for transfers, balances and standing orders. Which bias is likely?
1 answer is correct. Hard Created with AI
Show answer
Answer
- B Non-functional tests such as performance and security are underrepresented
Explanation
Functional flows are common in examples and training data, non-functional aspects less so. The model takes over this imbalance unless it is explicitly checked.