An AI model's validation report notes that the model performs well overall but fails on a specific edge case: when all optional fields in the input form are left blank. In an AI audit, how should this finding be classified?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Here's the deal — b is correct because known model failure modes — even on edge cases — must be documented as material limitations, and mitigating controls must be established (such as input validation to prevent blank submissions, or human override when blank inputs are detected). Undisclosed model limitations expose the organization to risk when those edge cases occur in production.
Full explanation below image
Full Explanation
B is correct because known model failure modes — even on edge cases — must be documented as material limitations, and mitigating controls must be established (such as input validation to prevent blank submissions, or human override when blank inputs are detected). Undisclosed model limitations expose the organization to risk when those edge cases occur in production. Dismissing it as minor (A) or acceptable (C) ignores the governance obligation to manage known failure modes. Attributing it to users (D) does not eliminate the organizational obligation to control the risk.