Microsoft Azure AI Fundamentals (AI-900) Practice Exam

What is a critical principle for responsible AI concerning the handling of predictions when essential fields have unusual or missing values?

1. Fairness

2. Accountability

3. Reliability and safety

4. Bias mitigation

The correct answer is: Reliability and safety

The principle of reliability and safety is crucial for responsible AI, especially when dealing with predictions that involve unusual or missing values in essential fields. This principle emphasizes the need for AI systems to produce consistent, accurate, and dependable outcomes, particularly in critical scenarios where the integrity of the predictions could lead to significant consequences. In practice, when an AI model encounters unusual or missing values, it is vital for the system to have mechanisms in place to handle these situations appropriately. This could involve implementing fallback strategies, providing alerts to users, or ensuring that the model’s reliability is maintained even when facing data irregularities. By prioritizing reliability and safety, organizations can enhance trust in their AI systems and ensure that decisions made based on AI predictions are sound and justified. Fairness focuses on eliminating bias and ensuring equitable outcomes for all users. Accountability refers to the responsibility of organizations and individuals for the decisions made by AI systems. Bias mitigation addresses the need to identify and reduce bias in AI models to promote fairness and equity. While all these principles are important for responsible AI, the specific context of handling predictions related to unusual or missing values aligns most closely with the principles of reliability and safety.