Free IBM watsonx Governance Lifecycle Advisor v1 - Associate Practice Questions

Sample Practice Questions

Try these 10 sample questions from the IBM watsonx Governance Lifecycle Advisor v1 - Associate practice bank. Each includes the answer and explanation. Continue with the interactive practice above to review the full question set.

1. 1Which IBM pillar of trustworthy AI focuses on equitable treatment of individuals or groups?

   1. A.FairnessCorrect answer: A
   2. B.Robustness
   3. C.Transparency
   4. D.Explainability

   Explanation: IBM states: "Fairness: This pillar refers to the equitable treatment of individuals, or groups of individuals, by an AI system." This identifies fairness as the relevant principle or concept.

   Reference: What is AI Ethics? — IBM’s point of view on AI ethics

2. 2A governance administrator wants to represent a bank's retail lending department in the governance console. Which object fits this organizational mapping?

   1. A.Model Deployment
   2. B.Model Risk Scorecard
   3. C.Business EntityCorrect answer: C
   4. D.Model Attestation

   Explanation: IBM's Redbook states: "For example: A bank's retail lending department is a Business Entity.". This supports the selected mapping or interaction mechanism.

   Reference: Ensuring Trustworthy AI with IBM watsonx.governance — Printed page 32, Business Entity

3. 3What does IBM's governance guidance mean by discovering shadow AI?

   1. A.Finding only models that have completed governance review
   2. B.Identifying models only after their retirement
   3. C.Finding AI already in use that has not been registeredCorrect answer: C
   4. D.Grouping registered models by deployment stage

   Explanation: IBM states: "Discovery of any AI that is already in use but not yet registered (shadow AI)." This supports the specified inventory decision.

   Reference: Ensuring Trustworthy AI with IBM watsonx.governance — 1.2.2 Concern 2: Governance does not scale

4. 4A team wants to compare different predictive frameworks for the same AI use case. What can it create within that use case?

   1. A.A separate inventory for every model within the use case
   2. B.A single approach with unrelated versions for each asset
   3. C.A separate regulatory mandate for each model
   4. D.Separate approaches within the use caseCorrect answer: D

   Explanation: IBM states: "For example, you might create two approaches to compare by using different frameworks for predictive models to see which one performs best." This supports the selected asset-tracking action or relationship.

   Reference: Tracking assets in an AI use case — Creating approaches to compare ways to solve a problem

5. 5When configuring generative AI quality evaluations, what distinguishes reference-free metrics from metrics that use feedback data?

   1. A.A reference input is required for every evaluation
   2. B.No reference input is required to calculate the metricCorrect answer: B
   3. C.A human-written reference is required at runtime
   4. D.A reference input is required for production data

   Explanation: IBM states: "there are reference free metrics as well, that do not require any reference input and can be calculated on the production data." This describes the selected evaluation capability or metric.

   Reference: Ensuring Trustworthy AI with IBM watsonx.governance — 6.2.4 Generative AI quality

6. 6Which IBM AI trust pillar addresses defense against adversarial attacks?

   1. A.Transparency
   2. B.RobustnessCorrect answer: B
   3. C.Privacy
   4. D.Explainability

   Explanation: IBM states: "Robustness: AI-powered systems must be actively defended from adversarial attacks, minimizing security risks and enabling confidence in system outcomes." This identifies robustness as the relevant principle or concept.

   Reference: What is AI Ethics? — IBM’s point of view on AI ethics

7. 7A bank wants to represent departments beneath a parent organizational unit in Governance console. How can Business Entities support this structure?

   1. A.Represent each department as a separate top-level-only entity
   2. B.Represent all departments with one unstructured entity
   3. C.Represent departments as attributes without sub-entities
   4. D.Represent departments as sub-entities of the parent entityCorrect answer: D

   Explanation: IBM states: "A business entity can contain sub-entities (such as departments, business units, or geographic locations)." The parent-and-sub-entity structure represents the hierarchy.

   Reference: Ensuring Trustworthy AI with IBM watsonx.governance — Printed page 32, Business Entity

8. 8When establishing an enterprise AI inventory, why integrate existing systems of record such as product catalogs?

   1. A.To repeat the same inventory gathering for each governance team
   2. B.To limit inventory data to the systems created by the AI team
   3. C.To avoid connecting existing records with AI governance
   4. D.To reuse information already captured in systems of recordCorrect answer: D

   Explanation: IBM states: "Use technology to leverage the information you already capture there so you do not have to redo that for governance purposes." This supports the specified inventory decision.

   Reference: Ensuring Trustworthy AI with IBM watsonx.governance — 1.2.2 Concern 2: Governance does not scale

9. 9How does watsonx.governance apply version information to assets within one approach?

   1. A.All assets in the approach share its version numberCorrect answer: A
   2. B.Each asset receives a separate unrelated approach version
   3. C.Only deployed assets receive the approach version
   4. D.Only the first tracked asset receives the approach version

   Explanation: IBM states: "The same version number is applied to all assets in an approach." This supports the selected asset-tracking action or relationship.

   Reference: Tracking assets in an AI use case — Creating approaches to compare ways to solve a problem

10. 10What does the LLM-as-a-judge feature allow in watsonx.governance?

    1. A.Use the evaluated LLM only to generate answers
    2. B.Use a human reviewer in place of an LLM evaluator
    3. C.Use an LLM to evaluate the performance of promptsCorrect answer: C
    4. D.Deploy the judging LLM as the target application model

    Explanation: IBM states: "IBM watsonx.governance also allows the use of widely available LLM models to evaluate the performance of the user prompts through the LLM-as-a-judge feature." This describes the selected evaluation capability or metric.

    Reference: Ensuring Trustworthy AI with IBM watsonx.governance — 6.2.4 Generative AI quality