Courseiva

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

What does the responsible AI principle of 'human in the loop' refer to?

⚠ Common exam trap

Candidates often confuse 'human in the loop' with general human involvement (like data entry or CAPTCHA) rather than recognizing it specifically as oversight of consequential AI decisions.

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

Maintaining human oversight and the ability to review or override consequential AI decisions

The 'human in the loop' principle ensures that humans maintain meaningful oversight over AI systems, particularly for high-stakes or consequential decisions. This means humans can review, override, or intervene in AI-generated outputs, preventing fully automated decision-making in critical scenarios such as medical diagnosis, loan approvals, or criminal justice. It is a core component of responsible AI, balancing automation with accountability.

Answer analysis

Option-by-option breakdown

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

  • A requirement for humans to manually enter all data into AI systems

    Why it's wrong here

    Requiring every data item to be manually entered shifts data preparation to humans, yet the AI system would still operate without human oversight at decision time. Manual entry is a pipeline workflow and often introduces transcription errors; it says nothing about who reviews or overrides the model's consequential conclusions. This misreads a responsibility principle as an input-handling chore.

  • Maintaining human oversight and the ability to review or override consequential AI decisions

    Why this is correct

    The correct definition: human-in-the-loop in responsible AI is a governance design that keeps humans accountable for high-stakes decisions, enabling them to review, approve, override, or reverse model outputs before or after they take effect. This control loop matters because models can be confidently wrong or operate in evolving contexts. Human oversight is proportionate to the decision's consequence level, from automated low-risk actions to mandatory review for irreversible actions.

  • Training AI models using feedback from human labelers only

    Why it's wrong here

    Reinforcement learning from human feedback is a training-time technique that uses labeler ratings to steer model behavior, but it does not guarantee that any human remains in control after deployment. Human-in-the-loop is broader, operationally grounded, and focuses on accountability for each consequential output. Training feedback and live oversight are different stages in the AI lifecycle; one does not substitute for the other.

  • Requiring users to prove they are human before using AI services

    Why it's wrong here

    CAPTCHA-based human verification is an access-control mechanism that blocks automated scripts, but it does not establish any ongoing human review of AI outputs. This option conflates user authentication with governance oversight. In responsible AI, human-in-the-loop specifically means a person can inspect, veto, or alter high-impact model decisions, not just prove a session is human.

About these practice questions

One of 985 original AI-900 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. 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.