CRISC Question #557: Real Exam Question with Answer & Explanation

Question

Which of the following is the GREATEST concern associated with the use of artificial intelligence (AI) language models?

Options

• AThe model could be hacked or exploited.
• BThe model could be used to generate inaccurate content.
• CStaff could become overly reliant on the model.
• DIt could lead to biased recommendations.

Explanation

The greatest concern with AI language models is their potential to perpetuate and amplify biases present in their training data, leading to unfair or discriminatory recommendations.

Common mistakes.

• A. While AI models can be hacked or exploited, this is a general security concern applicable to many systems, whereas bias is a more intrinsic and complex issue specific to the learning nature of AI.
• B. Generating inaccurate or fabricated content (hallucination) is a known issue, but it often can be addressed through human review and fact-checking, unlike the more subtle and pervasive issue of inherent bias.
• C. Staff becoming overly reliant on an AI model is a concern related to human-computer interaction, but it is secondary to the fundamental ethical problem of the model itself producing biased outputs.

Concept tested. AI ethical concerns (bias)

Reference. https://learn.microsoft.com/en-us/azure/applied-ai-services/azure-ai-content-safety/includes/responsible-ai-principles

No community discussion yet for this question.