Courseiva

AI-900 Practice Question: Describe Artificial Intelligence workloads and considerations

A hospital deploys an AI system to assist with diagnosing diseases from medical images. A doctor disagrees with the system's diagnosis and overrules it. The hospital wants to document this interaction for legal and audit purposes. Which Microsoft responsible AI principle is most directly relevant?

⚠ Common exam trap

Microsoft often tests the distinction between transparency (explaining how the AI works) and accountability (documenting who is responsible for decisions), leading candidates to incorrectly choose transparency when the question emphasizes legal documentation and audit trails.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

Accountability

The scenario involves documenting a human override of an AI system's diagnosis for legal and audit purposes, which directly relates to accountability. Accountability in responsible AI ensures that organizations can answer for their AI systems' decisions by maintaining clear records of interactions, including when humans overrule AI outputs. This principle requires traceability and governance mechanisms, such as audit trails, to assign responsibility for outcomes.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

  • Fairness

    Why it's wrong here

    Fairness in responsible AI is about preventing discriminatory outcomes and ensuring consistent treatment across demographic groups, often evaluated with metrics such as disparity in false-positive rates across protected classes. Documenting a clinician's manual override of an AI suggestion is an audit-trail action that records who changed the outcome and why; it does not measure or mitigate bias in model predictions. The override log could later support a fairness audit, but the act of documentation itself belongs to governance and accountability, not to fairness as a design principle.

  • Reliability and safety

    Why it's wrong here

    Reliability and safety require the AI system to perform robustly under expected clinical conditions, with verification, validation, and safeguards to prevent harm from incorrect outputs. Recording clinician overrides does not change model accuracy or establish system fail-safes; it preserves evidence of human intervention. While an override trail can reveal reliability issues (e.g., frequent overrides suggest poor model calibration), the practice of maintaining such records is fundamentally an accountability mechanism, because it assigns responsibility for decisions.

  • Transparency

    Why it's wrong here

    Transparency is about making the AI system's behavior understandable—through model cards, feature-importance explanations, or plain-language descriptions of how inputs produce outputs—so clinicians can interpret a recommendation. Documenting a human override is not a disclosure of model logic; it is a procedural record that identifies who took responsibility and what action was taken. Transparency answers 'how did the model arrive at this output?', whereas documentation of overrides answers 'who is accountable for the final clinical decision?'

  • Accountability

    Why this is correct

    Accountability is the responsible AI principle that assigns clear human ownership and oversight for system outcomes, requiring role definition, audit trails, and records of decisions, including instances where a clinician overrides an AI recommendation. In healthcare, an override log enables retrospective review, incident analysis, and regulatory compliance by showing that a qualified human retained ultimate responsibility. This principle directly matches the scenario: documenting overrides creates the traceability needed to hold individuals and institutions answerable.

About these practice questions

This AI-900 question is part of Courseiva's 985-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →

How Courseiva writes practice questions · Editorial policy

JA

Written by Johnson Ajibi, MSc IT Security

Senior Network & Security Engineer · founder of Courseiva

This AI-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-900 exam.