Domain 2: Generative AI Ethics and Bias (25%) Topics: Analyze ethics and bias in generative AI, Develop strategies to mitigate bias in generative AI

Domain 3: Generative AI Strategy and Leadership (15%) Topics: Develop a generative AI strategy and roadmap, Lead and govern generative AI projects and initiatives

Domain 4: Generative AI Implementation and Governance (20%) Topics: Implement and govern generative AI solutions in an organization, Monitor and evaluate the effectiveness of generative AI solutions

Domain 5: Generative AI Security and Risk Management (20%) Topics: Assess and mitigate security and risk associated with generative AI, Develop incident response and crisis management strategies for generative AI

B) It provides 24/7 personalized responses at scale but introduces the risk of hallucinating incorrect information.

B) It guarantees 100% accuracy in technical troubleshooting but is universally too expensive for enterprise use.

D) It securely stores all customer passwords but significantly slows down the average handling time.

E) It automatically resolves all physical hardware defects but struggles with basic software issues.

F) It replaces the entire CRM architecture but requires isolated physical servers on-premise.

Correct Answer: B

Explanation:

Option B is correct because generative AI excels at scaling personalized communication around the clock, but hallucination (generating false or nonsensical information) is a known limitation that requires human oversight and governance.

Option C is incorrect because generative AI does not guarantee 100% accuracy, and costs can be managed depending on the deployment model.

Option D is incorrect because AI should not be used as a primary secure storage vault for passwords, and it generally speeds up handling time rather than slowing it down.

Option E is incorrect because generative AI is a software solution and cannot physically repair hardware defects.

Option F is incorrect because generative AI integrates with CRM systems rather than replacing their foundational architecture.

Question 2: When developing a strategy to mitigate bias in a generative AI deployment, which of the following actions should a leader prioritize during the initial governance phase?

B) Exclusively use publicly available internet data without filtering to ensure natural diversity.

C) Establish a diverse human-in-the-loop review board to systematically evaluate training data and model outputs.

D) Limit model access to only the executive management team to prevent external PR issues.

E) Hardcode specific demographic outputs to force statistical parity regardless of the conversational context.

F) Outsource all ethics compliance to a generic third-party vendor without maintaining internal oversight.

Correct Answer: C

Explanation:

Option B is incorrect because unfiltered internet data is historically filled with biases and toxic content, which the model will inherit.

Option C is correct because a diverse, cross-functional review board ensures that multiple perspectives are considered when evaluating data sets and model behavior, which is a core tenet of responsible AI governance.

Option D is incorrect because limiting access to executives does not solve the underlying bias problem; it merely hides it.

Option E is incorrect because forcing hardcoded responses ruins the contextual relevance of generative AI and can introduce new forms of bias.

Option F is incorrect because while third parties can help, an organization's leadership must maintain ultimate accountability and oversight for AI ethics.

Question 3: A generative AI system deployed internally is found to be leaking sensitive proprietary code in its prompt responses. What is the most appropriate immediate incident response strategy?

B) Isolate the AI system from the internal network, revoke API access, and initiate a forensic audit.

C) Issue a public press release apologizing for the data leak before conducting any internal investigation.

D) Increase the model's temperature parameter to encourage it to generate different, non-sensitive responses.

E) Delete all company training data permanently and purchase a new off-the-shelf commercial model.

F) Ignore the issue temporarily if the leaked proprietary code is older than three years.

Correct Answer: B

Explanation:

Option B is correct because the immediate priority in any security breach is containment. Isolating the system prevents further leaks while a forensic audit determines the root cause.

Option C is incorrect because issuing public statements before understanding the scope of an internal leak can cause unnecessary panic and misinformation.

Option D is incorrect because altering the temperature parameter only changes the randomness of the output and does not patch the underlying data leakage vulnerability.

Option E is incorrect because permanently deleting data destroys evidence needed for the audit and causes massive business disruption.

Option F is incorrect because proprietary code, regardless of age, poses a security risk and ignoring a data leak violates basic risk management principles.

Welcome to the Mock Exam Practice Tests Academy to help you prepare for your Generative AI Leader Certification.

You can retake the exams as many times as you want

This is a huge original question bank

You get support from instructors if you have questions

Each question has a detailed explanation

Mobile-compatible with the Udemy app