Question 1080An LLM assistant's responses become less grounded after a knowledge index update, while latency looks healthy. What controlled validation most directly confirms the approach worked? Apply a safe controlled load and verify p95 latency, error-rate and resource alarms trigger with clear ownership. Test a suite of adversarial and unauthorized prompts and verify sensitive output is blocked and auditable. Compare sampled source citations and factuality against approved references before and after the index update. Compare recent labeled metrics and calibration to a frozen prior period under comparable evaluation rules. ExplanationCorrect Answer C:Compare sampled source citations and factuality against approved references before and after the index update. This tests the underlying failure: Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. Other Answers A:Reliability metrics expose serving failures independently from predictive model quality and training drift. That does not establish the required outcome; verify it by checking: Compare sampled source citations and factuality against approved references before and after the index update. 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 sampled source citations and factuality against approved references before and after the index update. D: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 sampled source citations and factuality against approved references before and after the index update.
Question 1081An LLM assistant's responses become less grounded after a knowledge index update, while latency looks healthy. What preventive release checklist item best protects the intended outcome? Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. Maintain policy test suites and review false refusals whenever model and safety-filter versions change. Define attribution-change alert policies and require domain analysis before model replacement. Refresh reference distributions only after approved investigation and maintain source-drift incident runbooks. ExplanationCorrect Answer A:Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. This protects the selected approach because Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. The immediate solution is: Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. Other Answers B:Model instructions are probabilistic and safety guardrails need separate policy checks at input and output boundaries. The continuing safeguard required here is: Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. C:Attribution drift can signal changes in how a model uses information that overall aggregate metrics may miss. The continuing safeguard required here is: Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. D: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 retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests.
Question 1082A document-answering agent receives new policy pages and starts citing irrelevant sources despite unchanged prompts. What should the team change to address the specific technical requirement? Restrict the workload service account to least-privilege roles and the minimum approved resource set. Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. Configure Agent Platform Model Monitoring to alert on meaningful feature-distribution drift from a valid reference dataset. Add training-serving feature-skew checks and deploy the exact versioned preprocessing artifacts used during training. ExplanationCorrect Answer B:Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. To validate the correction: Compare sampled source citations and factuality against approved references before and after the index update. 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: Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. C:Data drift is a change in input distributions and can occur even when prediction-outcome labels are delayed. It does not resolve the specific failure; the targeted correction is: Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. D:Training-serving skew can degrade predictions without any change to the underlying fitted model weights. It does not resolve the specific failure; the targeted correction is: Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset.
Question 1083A document-answering agent receives new policy pages and starts citing irrelevant sources despite unchanged prompts. Which property of the selected ML workflow accounts for this requirement? Attribution drift can signal changes in how a model uses information that overall aggregate metrics may miss. Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. Reliability metrics expose serving failures independently from predictive model quality and training drift. ExplanationCorrect Answer B:Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. This supports the specific implementation: Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. A suitable independent check is: Compare sampled source citations and factuality against approved references before and after the index update. Other Answers A:Attribution drift can signal changes in how a model uses information that overall aggregate metrics may miss. The cause in the stated scenario is different: Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. C:Feature attribution describes how inputs relate to predictions, but explanations are not necessarily causal proof. The cause in the stated scenario is different: Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. D:Reliability metrics expose serving failures independently from predictive model quality and training drift. The cause in the stated scenario is different: Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy.
Question 1084A document-answering agent receives new policy pages and starts citing irrelevant sources despite unchanged prompts. Which evidence would best demonstrate that this ML solution meets the requirement? Compare sampled source citations and factuality against approved references before and after the index update. Compare recent labeled metrics and calibration to a frozen prior period under comparable evaluation rules. Submit defined unsafe test inputs and inspect filter decisions, false positives and user-safe fallback responses. Replay shifted fixture data and confirm configured drift thresholds and alert routing work as expected. ExplanationCorrect Answer A:Compare sampled source citations and factuality against approved references before and after the index update. This tests the underlying failure: Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. Other Answers B: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 sampled source citations and factuality against approved references before and after the index update. C:Model instructions are probabilistic and safety guardrails need separate policy checks at input and output boundaries. That does not establish the required outcome; verify it by checking: Compare sampled source citations and factuality against approved references before and after the index update. 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 sampled source citations and factuality against approved references before and after the index update.
Question 1085A document-answering agent receives new policy pages and starts citing irrelevant sources despite unchanged prompts. Which ongoing standard would make this result reliable for future model versions? Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. Refresh reference distributions only after approved investigation and maintain source-drift incident runbooks. Require periodic IAM entitlement recertification and automatic policy checks before model deployment. Maintain separate service reliability and model-quality SLOs with tested incident escalation paths. ExplanationCorrect Answer A:Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. This protects the selected approach because Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. The immediate solution is: Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. Other Answers B: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 retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. C:IAM controls cloud API permissions independently of whether a model produces accurate or safe predictions. The continuing safeguard required here is: Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. D:Reliability metrics expose serving failures independently from predictive model quality and training drift. The continuing safeguard required here is: Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests.
Question 1086A retrieval-augmented assistant shows more unsupported assertions following a retriever configuration change. Which Google Cloud machine-learning design choice best satisfies the stated requirement? Monitor labeled performance and investigate concept drift using refreshed ground truth instead of input drift alone. Enforce retrieval access controls and layered safety inspection such as Model Armor for malicious prompts and sensitive outputs. Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. Add training-serving feature-skew checks and deploy the exact versioned preprocessing artifacts used during training. ExplanationCorrect Answer C:Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. To validate the correction: Compare sampled source citations and factuality against approved references before and after the index update. Other Answers A:Concept drift changes the relationship between features and targets rather than necessarily changing feature marginals. It does not resolve the specific failure; the targeted correction is: Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. B: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: Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. D:Training-serving skew can degrade predictions without any change to the underlying fitted model weights. It does not resolve the specific failure; the targeted correction is: Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset.
Question 1087A retrieval-augmented assistant shows more unsupported assertions following a retriever configuration change. Which technical mechanism is most relevant to this outcome? Data drift is a change in input distributions and can occur even when prediction-outcome labels are delayed. Attribution drift can signal changes in how a model uses information that overall aggregate metrics may miss. Aggregate accuracy can obscure uneven subgroup errors and biased training labels or sampling distributions. Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. ExplanationCorrect Answer D:Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. This supports the specific implementation: Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. A suitable independent check is: Compare sampled source citations and factuality against approved references before and after the index update. 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: Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. B:Attribution drift can signal changes in how a model uses information that overall aggregate metrics may miss. The cause in the stated scenario is different: Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. C:Aggregate accuracy can obscure uneven subgroup errors and biased training labels or sampling distributions. The cause in the stated scenario is different: Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy.
Question 1088A retrieval-augmented assistant shows more unsupported assertions following a retriever configuration change. What controlled validation most directly confirms the approach worked? Compare sampled source citations and factuality against approved references before and after the index update. Apply a safe controlled load and verify p95 latency, error-rate and resource alarms trigger with clear ownership. Run allowed and denied access tests using the actual serving identity and audit role grants. Replay identical raw cases in both environments and compare feature vectors and model outputs. ExplanationCorrect Answer A:Compare sampled source citations and factuality against approved references before and after the index update. This tests the underlying failure: Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. Other Answers B:Reliability metrics expose serving failures independently from predictive model quality and training drift. That does not establish the required outcome; verify it by checking: Compare sampled source citations and factuality against approved references before and after the index update. 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 sampled source citations and factuality against approved references before and after the index update. D: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 sampled source citations and factuality against approved references before and after the index update.
Question 1089A retrieval-augmented assistant shows more unsupported assertions following a retriever configuration change. Which ongoing standard would make this result reliable for future model versions? Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. Require periodic IAM entitlement recertification and automatic policy checks before model deployment. Define attribution-change alert policies and require domain analysis before model replacement. Refresh reference distributions only after approved investigation and maintain source-drift incident runbooks. ExplanationCorrect Answer A:Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. This protects the selected approach because Generative AI answer quality can regress through retrieval or prompt changes even when infrastructure is healthy. The immediate solution is: Run continuous groundedness, citation relevance and task-completion evaluations on a held-out generative AI dataset. Other Answers B:IAM controls cloud API permissions independently of whether a model produces accurate or safe predictions. The continuing safeguard required here is: Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. C:Attribution drift can signal changes in how a model uses information that overall aggregate metrics may miss. The continuing safeguard required here is: Version retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests. D: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 retrieval indexes and evaluation prompts and gate releases on grounded-answer regression tests.
Question 1090A model service exceeds its inference latency SLO but current dashboards display only validation accuracy. What should the team change to address the specific technical requirement? Enforce retrieval access controls and layered safety inspection such as Model Armor for malicious prompts and sensitive outputs. Instrument request latency, availability, utilization and queueing with production Model Monitoring and cloud telemetry. Add training-serving feature-skew checks and deploy the exact versioned preprocessing artifacts used during training. Configure Agent Platform Model Monitoring to alert on meaningful feature-distribution drift from a valid reference dataset. ExplanationCorrect Answer B:Reliability metrics expose serving failures independently from predictive model quality and training drift. To validate the correction: Apply a safe controlled load and verify p95 latency, error-rate and resource alarms trigger with clear ownership. 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: Instrument request latency, availability, utilization and queueing with production Model Monitoring and cloud telemetry. C:Training-serving skew can degrade predictions without any change to the underlying fitted model weights. It does not resolve the specific failure; the targeted correction is: Instrument request latency, availability, utilization and queueing with production Model Monitoring and cloud telemetry. D:Data drift is a change in input distributions and can occur even when prediction-outcome labels are delayed. It does not resolve the specific failure; the targeted correction is: Instrument request latency, availability, utilization and queueing with production Model Monitoring and cloud telemetry.