Which of the following AI system characteristics would BEST help an IS auditor evaluate the system's algorithm?

Question

Options

• AThe AI system algorithm uses training data to inform decision output.
• BThe AI system provides multiple options for model training.
• CThe AI system provides transparent justification of decisions.
• DThe AI system uses archived transaction data to provide decisions.

• A
  13% (6)
• B
  4% (2)
• C
  78% (35)
• D
  4% (2)

Explanation

An IS auditor evaluating an algorithm needs to understand the reasoning behind its decisions-not just what decisions it makes. A system that provides transparent justification of decisions (C) allows the auditor to assess whether the logic is sound, fair, consistent, and appropriate for the use case. This is the core of algorithm auditability. That a model uses training data (A) is a basic, universal characteristic that does not help the auditor evaluate its logic. Providing multiple model training options (B) is a design feature, not an auditability characteristic. Using archived transaction data (D) describes the data source, not the decision-making process itself.

No community discussion yet for this question.