AAISM: Advanced in AI Security Management

By Isaca · Question Mode

QuestionQ65

VerifiedAI Risk Management

An organization has recently deployed a generative AI chatbot that interacts with users and responds to their queries. Which of the following would BEST mitigate the hallucination risk identified by the risk team?

Explanation

Model testing and validation directly assess whether the chatbot produces inaccurate or fabricated responses, enabling the organization to identify, measure, and remediate hallucination behavior before or during operation.

Community Discussion

No comments yet. Be the first to start the discussion!