Skip to content

Quiz 8: Governance, Ethics & Responsible AI

Test your understanding of AI governance, ethics, and responsible deployment.

Questions

Question 1 (Remember)

What is a GAI Center of Excellence (GAICoE)?

  1. An AI model
  2. An organizational unit for developing and scaling AI capabilities
  3. A regulatory body
  4. A type of API
Answer

B) An organizational unit for developing and scaling AI capabilities - A GAICoE provides governance, best practices, training, and support for AI initiatives across the organization.

Question 2 (Remember)

What is the EU AI Act?

  1. A US regulation
  2. An AI model
  3. European legislation establishing rules for AI development and use
  4. A private industry standard
Answer

C) European legislation establishing rules for AI development and use - The EU AI Act is a risk-based regulatory framework governing AI systems in the European Union.

Question 3 (Understand)

What is the purpose of red-teaming in AI systems?

  1. To make AI faster
  2. To find vulnerabilities and failure modes through adversarial testing
  3. To train new models
  4. To reduce costs
Answer

B) To find vulnerabilities and failure modes through adversarial testing - Red teams deliberately try to "break" AI systems to identify weaknesses, biases, and safety issues before deployment.

Question 4 (Understand)

Why is AI transparency important?

  1. It makes AI faster
  2. It reduces development costs
  3. It enables understanding of AI decisions and builds trust
  4. It is only relevant for open-source models
Answer

C) It enables understanding of AI decisions and builds trust - Transparency helps users and stakeholders understand how AI systems work, their limitations, and the basis for their outputs.

Question 5 (Apply)

Your organization is implementing AI for hiring decisions. What governance measure is most critical?

  1. Faster processing
  2. Cost reduction
  3. Bias detection and mitigation with human oversight
  4. Automated decision-making without review
Answer

C) Bias detection and mitigation with human oversight - High-stakes decisions like hiring require careful attention to fairness, with systems to detect bias, mitigation strategies, and human review of AI recommendations.

Question 6 (Apply)

A customer asks how your AI chatbot made a recommendation. What principle are they invoking?

  1. Efficiency
  2. Explainability
  3. Speed
  4. Creativity
Answer

B) Explainability - Explainability is the ability to describe AI decision-making in understandable terms. Users have a reasonable expectation to understand significant AI-driven recommendations.

Question 7 (Analyze)

Compare rule-based AI governance with principles-based governance:

  1. They are identical
  2. Rules are specific and prescriptive; principles provide flexible guidance
  3. Principles are more restrictive
  4. Rules are never used
Answer

B) Rules are specific and prescriptive; principles provide flexible guidance - Rules define specific requirements (do X, don't do Y), while principles establish values and goals that guide decisions across varied situations.

Question 8 (Analyze)

What is the relationship between AI hallucinations and responsible AI deployment?

  1. Hallucinations are desirable
  2. They are unrelated
  3. Managing hallucination risk is a key responsible AI requirement
  4. Hallucinations only affect images
Answer

C) Managing hallucination risk is a key responsible AI requirement - Responsible AI requires acknowledging that models can generate incorrect information and implementing safeguards (grounding, verification, disclaimers).

Question 9 (Evaluate)

An AI system shows higher error rates for certain demographic groups. How should this be addressed?

  1. Ignore it if overall accuracy is good
  2. Remove the demographic data
  3. Investigate root causes, implement bias mitigation, and monitor continuously
  4. Deploy anyway and fix later
Answer

C) Investigate root causes, implement bias mitigation, and monitor continuously - Disparate performance requires understanding why it occurs, taking corrective action, and ongoing monitoring to ensure the issue is resolved.

Question 10 (Create)

Design a GAICoE charter for a healthcare organization with strict regulatory requirements and diverse AI use cases.

  1. Copy a generic template
  2. Focus only on technology
  3. Include: mission, scope, governance structure, compliance framework, ethics review, stakeholder engagement, success metrics
  4. Wait for regulations to be finalized
Answer

C) Include: mission, scope, governance structure, compliance framework, ethics review, stakeholder engagement, success metrics - Healthcare GAICoE needs comprehensive governance addressing the unique regulatory (HIPAA), ethical (patient safety), and organizational requirements.

Score Interpretation

  • 9-10 correct: Excellent understanding of AI governance and ethics
  • 7-8 correct: Good grasp, review missed concepts
  • 5-6 correct: Fair understanding, revisit chapter sections
  • Below 5: Re-read Chapter 8 before proceeding

Previous: Quiz 7 | Back to Quizzes | Next: Quiz 9 - Future of Work