Question 980A public services classifier needs evidence of fairness and disparate-impact monitoring before expanding use. Which Google Cloud machine-learning design choice best satisfies the stated requirement? Enforce retrieval access controls and layered safety inspection such as Model Armor for malicious prompts and sensitive outputs. Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations. Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. Evaluate model outcomes and error rates across relevant groups and apply documented responsible-AI mitigations. ExplanationCorrect Answer D:Aggregate accuracy can obscure uneven subgroup errors and biased training labels or sampling distributions. To validate the correction: Compare group metrics and representative false negatives on consented and legally appropriate audit samples. Other Answers A:Prompt injection is untrusted input attempting to override application instructions and can cause data leakage without isolation. It does not resolve the specific failure; the targeted correction is: Evaluate model outcomes and error rates across relevant groups and apply documented responsible-AI mitigations. B:Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. It does not resolve the specific failure; the targeted correction is: Evaluate model outcomes and error rates across relevant groups and apply documented responsible-AI mitigations. C:Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. It does not resolve the specific failure; the targeted correction is: Evaluate model outcomes and error rates across relevant groups and apply documented responsible-AI mitigations.
Question 981A public services classifier needs evidence of fairness and disparate-impact monitoring before expanding use. Which technical mechanism is most relevant to this outcome? Model instructions are probabilistic and safety guardrails need separate policy checks at input and output boundaries. Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. Data drift is a change in input distributions and can occur even when prediction-outcome labels are delayed. Aggregate accuracy can obscure uneven subgroup errors and biased training labels or sampling distributions. ExplanationCorrect Answer D:Aggregate accuracy can obscure uneven subgroup errors and biased training labels or sampling distributions. This supports the specific implementation: Evaluate model outcomes and error rates across relevant groups and apply documented responsible-AI mitigations. A suitable independent check is: Compare group metrics and representative false negatives on consented and legally appropriate audit samples. Other Answers A:Model instructions are probabilistic and safety guardrails need separate policy checks at input and output boundaries. The cause in the stated scenario is different: Aggregate accuracy can obscure uneven subgroup errors and biased training labels or sampling distributions. B:Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. The cause in the stated scenario is different: Aggregate accuracy can obscure uneven subgroup errors and biased training labels or sampling distributions. C:Data drift is a change in input distributions and can occur even when prediction-outcome labels are delayed. The cause in the stated scenario is different: Aggregate accuracy can obscure uneven subgroup errors and biased training labels or sampling distributions.
Question 982A public services classifier needs evidence of fairness and disparate-impact monitoring before expanding use. What is the most relevant acceptance check for the proposed ML change? Replay identical raw cases in both environments and compare feature vectors and model outputs. Inject synthetic secrets and verify redaction, log access denials and retention configuration. Run allowed and denied access tests using the actual serving identity and audit role grants. Compare group metrics and representative false negatives on consented and legally appropriate audit samples. ExplanationCorrect Answer D:Compare group metrics and representative false negatives on consented and legally appropriate audit samples. This tests the underlying failure: Aggregate accuracy can obscure uneven subgroup errors and biased training labels or sampling distributions. Other Answers A:Training-serving skew can degrade predictions without any change to the underlying fitted model weights. That does not establish the required outcome; verify it by checking: Compare group metrics and representative false negatives on consented and legally appropriate audit samples. B:Observability data can itself expose confidential training or inference content if logs are collected without safeguards. That does not establish the required outcome; verify it by checking: Compare group metrics and representative false negatives on consented and legally appropriate audit samples. C:IAM controls cloud API permissions independently of whether a model produces accurate or safe predictions. That does not establish the required outcome; verify it by checking: Compare group metrics and representative false negatives on consented and legally appropriate audit samples.
Question 983A public services classifier needs evidence of fairness and disparate-impact monitoring before expanding use. Which ongoing standard would make this result reliable for future model versions? Require fairness monitoring, documented mitigation ownership and human escalation for high-impact decisions. Require periodic IAM entitlement recertification and automatic policy checks before model deployment. Maintain policy test suites and review false refusals whenever model and safety-filter versions change. Maintain separate service reliability and model-quality SLOs with tested incident escalation paths. ExplanationCorrect Answer A:Require fairness monitoring, documented mitigation ownership and human escalation for high-impact decisions. This protects the selected approach because Aggregate accuracy can obscure uneven subgroup errors and biased training labels or sampling distributions. The immediate solution is: Evaluate model outcomes and error rates across relevant groups and apply documented responsible-AI mitigations. Other Answers B:IAM controls cloud API permissions independently of whether a model produces accurate or safe predictions. The continuing safeguard required here is: Require fairness monitoring, documented mitigation ownership and human escalation for high-impact decisions. C:Model instructions are probabilistic and safety guardrails need separate policy checks at input and output boundaries. The continuing safeguard required here is: Require fairness monitoring, documented mitigation ownership and human escalation for high-impact decisions. D:Reliability metrics expose serving failures independently from predictive model quality and training drift. The continuing safeguard required here is: Require fairness monitoring, documented mitigation ownership and human escalation for high-impact decisions.
Question 984A regulated scoring model must provide a defensible explanation of which inputs affected individual decisions. Which proposed approach best resolves the described model engineering problem? Evaluate model outcomes and error rates across relevant groups and apply documented responsible-AI mitigations. Restrict the workload service account to least-privilege roles and the minimum approved resource set. Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations. Monitor feature-attribution drift alongside outcome and input metrics on representative authorized samples. ExplanationCorrect Answer C:Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. To validate the correction: Compare local attribution stability on similar cases and check against controlled perturbations and business rules. Other Answers A:Aggregate accuracy can obscure uneven subgroup errors and biased training labels or sampling distributions. It does not resolve the specific failure; the targeted correction is: Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations. B:IAM controls cloud API permissions independently of whether a model produces accurate or safe predictions. It does not resolve the specific failure; the targeted correction is: Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations. D:Attribution drift can signal changes in how a model uses information that overall aggregate metrics may miss. It does not resolve the specific failure; the targeted correction is: Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations.
Question 985A regulated scoring model must provide a defensible explanation of which inputs affected individual decisions. Which property of the selected ML workflow accounts for this requirement? Data drift is a change in input distributions and can occur even when prediction-outcome labels are delayed. IAM controls cloud API permissions independently of whether a model produces accurate or safe predictions. Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. Model instructions are probabilistic and safety guardrails need separate policy checks at input and output boundaries. ExplanationCorrect Answer C:Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. This supports the specific implementation: Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations. A suitable independent check is: Compare local attribution stability on similar cases and check against controlled perturbations and business rules. Other Answers A:Data drift is a change in input distributions and can occur even when prediction-outcome labels are delayed. The cause in the stated scenario is different: Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. B:IAM controls cloud API permissions independently of whether a model produces accurate or safe predictions. The cause in the stated scenario is different: Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. D:Model instructions are probabilistic and safety guardrails need separate policy checks at input and output boundaries. The cause in the stated scenario is different: Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof.
Question 986A regulated scoring model must provide a defensible explanation of which inputs affected individual decisions. Which evidence would best demonstrate that this ML solution meets the requirement? Compare recent labeled metrics and calibration to a frozen prior period under comparable evaluation rules. Test a suite of adversarial and unauthorized prompts and verify sensitive output is blocked and auditable. Inject synthetic secrets and verify redaction, log access denials and retention configuration. Compare local attribution stability on similar cases and check against controlled perturbations and business rules. ExplanationCorrect Answer D:Compare local attribution stability on similar cases and check against controlled perturbations and business rules. This tests the underlying failure: Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. Other Answers A:Concept drift changes the relationship between features and targets rather than necessarily changing feature marginals. That does not establish the required outcome; verify it by checking: Compare local attribution stability on similar cases and check against controlled perturbations and business rules. B:Prompt injection is untrusted input attempting to override application instructions and can cause data leakage without isolation. That does not establish the required outcome; verify it by checking: Compare local attribution stability on similar cases and check against controlled perturbations and business rules. C:Observability data can itself expose confidential training or inference content if logs are collected without safeguards. That does not establish the required outcome; verify it by checking: Compare local attribution stability on similar cases and check against controlled perturbations and business rules.
Question 987A regulated scoring model must provide a defensible explanation of which inputs affected individual decisions. What governance or operational safeguard should accompany the implementation? Refresh reference distributions only after approved investigation and maintain source-drift incident runbooks. Version explainability methodology and require explanation quality reviews when features or architectures change. Maintain prompt-injection regression tests, least-privilege retrieval and incident review for new knowledge sources. Maintain policy test suites and review false refusals whenever model and safety-filter versions change. ExplanationCorrect Answer B:Version explainability methodology and require explanation quality reviews when features or architectures change. This protects the selected approach because Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. The immediate solution is: Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations. Other Answers A:Data drift is a change in input distributions and can occur even when prediction-outcome labels are delayed. The continuing safeguard required here is: Version explainability methodology and require explanation quality reviews when features or architectures change. C:Prompt injection is untrusted input attempting to override application instructions and can cause data leakage without isolation. The continuing safeguard required here is: Version explainability methodology and require explanation quality reviews when features or architectures change. D:Model instructions are probabilistic and safety guardrails need separate policy checks at input and output boundaries. The continuing safeguard required here is: Version explainability methodology and require explanation quality reviews when features or architectures change.
Question 988A mortgage operations team needs feature attribution evidence for an approved risk model. Which implementation response is most appropriate for this ML engineering situation? Restrict the workload service account to least-privilege roles and the minimum approved resource set. Monitor feature-attribution drift alongside outcome and input metrics on representative authorized samples. Use supported safety filters and explicit application-level response validation rather than relying on prompts alone. Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations. ExplanationCorrect Answer D:Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. To validate the correction: Compare local attribution stability on similar cases and check against controlled perturbations and business rules. Other Answers A:IAM controls cloud API permissions independently of whether a model produces accurate or safe predictions. It does not resolve the specific failure; the targeted correction is: Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations. B:Attribution drift can signal changes in how a model uses information that overall aggregate metrics may miss. It does not resolve the specific failure; the targeted correction is: Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations. C:Model instructions are probabilistic and safety guardrails need separate policy checks at input and output boundaries. It does not resolve the specific failure; the targeted correction is: Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations.
Question 989A mortgage operations team needs feature attribution evidence for an approved risk model. Which property of the selected ML workflow accounts for this requirement? IAM controls cloud API permissions independently of whether a model produces accurate or safe predictions. Observability data can itself expose confidential training or inference content if logs are collected without safeguards. Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. Data drift is a change in input distributions and can occur even when prediction-outcome labels are delayed. ExplanationCorrect Answer C:Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. This supports the specific implementation: Use supported model-explainability methods on Agent Platform and validate attribution against domain expectations. A suitable independent check is: Compare local attribution stability on similar cases and check against controlled perturbations and business rules. Other Answers A:IAM controls cloud API permissions independently of whether a model produces accurate or safe predictions. The cause in the stated scenario is different: Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. B:Observability data can itself expose confidential training or inference content if logs are collected without safeguards. The cause in the stated scenario is different: Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. D:Data drift is a change in input distributions and can occur even when prediction-outcome labels are delayed. The cause in the stated scenario is different: Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof.
Question 990A mortgage operations team needs feature attribution evidence for an approved risk model. What controlled validation most directly confirms the approach worked? Run allowed and denied access tests using the actual serving identity and audit role grants. Test a suite of adversarial and unauthorized prompts and verify sensitive output is blocked and auditable. Compare local attribution stability on similar cases and check against controlled perturbations and business rules. Replay shifted fixture data and confirm configured drift thresholds and alert routing work as expected. ExplanationCorrect Answer C:Compare local attribution stability on similar cases and check against controlled perturbations and business rules. This tests the underlying failure: Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. Other Answers A:IAM controls cloud API permissions independently of whether a model produces accurate or safe predictions. That does not establish the required outcome; verify it by checking: Compare local attribution stability on similar cases and check against controlled perturbations and business rules. B:Prompt injection is untrusted input attempting to override application instructions and can cause data leakage without isolation. That does not establish the required outcome; verify it by checking: Compare local attribution stability on similar cases and check against controlled perturbations and business rules. D:Data drift is a change in input distributions and can occur even when prediction-outcome labels are delayed. That does not establish the required outcome; verify it by checking: Compare local attribution stability on similar cases and check against controlled perturbations and business rules.