McAfee-Secured Website

Exam Code: AI-300

Exam Name: Operationalizing Machine Learning and Generative AI Solutions

Certification Provider: Microsoft

Corresponding Certification: Microsoft Certified: Machine Learning Operations Engineer Associate

100% Microsoft AI-300 Guaranteed Success With Testking Training Materials

Get Real AI-300 Exam Questions & Expert Verified Answers as Experienced on the Real Exam!

213 Questions & Answers with Testing Engine

"Operationalizing Machine Learning and Generative AI Solutions Exam", also known as AI-300 exam, is a Microsoft certification exam.

Pass your tests with the always up-to-date AI-300 Exam Engine. Your AI-300 training materials keep you at the head of the pack!

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Was: $137.49
Now: $124.99

Microsoft AI-300 Practice Test 31

Question 300

A decommissioned application no longer uses a production model, and storage costs grow with obsolete artifacts. What cleanup strategy is appropriate?

  1. Archive or remove obsolete model versions only after dependency and retention checks.
  2. Maintain accountable owner and lifecycle status metadata for every registered model family.
  3. Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state.
  4. Use a stable registered model name and explicit version rather than transient storage paths.

Explanation

Correct Answer

A:

Safe lifecycle cleanup balances operational cost with rollback needs and legally required artifact retention.

Incorrect Answers

B:

Maintain accountable owner and lifecycle status metadata for every registered model family. This is useful for the distinct issue: A data science team has no record of who owns models abandoned in a shared registry. The relevant technical rationale or evidence for that issue is: Ownership information enables safe review, maintenance and retirement decisions when model-producing staff or projects change. Here the reported issue is: A decommissioned application no longer uses a production model, and storage costs grow with obsolete artifacts. The relevant response is: Archive or remove obsolete model versions only after dependency and retention checks. The first result cannot verify this different correction: Safe lifecycle cleanup balances operational cost with rollback needs and legally required artifact retention.

C:

Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state. It addresses another case: A team needs to show why an older model version stopped serving due to fairness concerns. For that case, the distinguishing justification or acceptance check is: Governance metadata explains retirement decisions and prevents unsafe unreviewed reactivation of an old artifact. The present case instead states: A decommissioned application no longer uses a production model, and storage costs grow with obsolete artifacts. It calls for Archive or remove obsolete model versions only after dependency and retention checks. Showing success for the other case would not establish: Safe lifecycle cleanup balances operational cost with rollback needs and legally required artifact retention.

D:

The alternative, Use a stable registered model name and explicit version rather than transient storage paths, belongs to a different technical condition: A production rollout references a model by its environment-specific storage URI, which changes after a workspace migration. There the source-specific mechanism or proof is: Registry identifiers separate logical model identity from storage location, improving portability across controlled deployments. The condition asked about here is: A decommissioned application no longer uses a production model, and storage costs grow with obsolete artifacts. A suitable correction is Archive or remove obsolete model versions only after dependency and retention checks; the targeted justification or result is: Safe lifecycle cleanup balances operational cost with rollback needs and legally required artifact retention.

Question 301

A training team creates a better model but reuses the prior version number in deployment notes. How should release traceability work?

  1. Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state.
  2. Record immutable review outcomes and approval timestamps with the model release metadata.
  3. Retain or document the feature definitions and dependencies required to reproduce the archived model.
  4. Assign and reference the distinct new registered model version in deployment records.

Explanation

Correct Answer

D:

Version references keep the deployed artifact unambiguous and preserve the history of successive model promotions.

Incorrect Answers

A:

Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state. A separate source case supports that approach when: A team needs to show why an older model version stopped serving due to fairness concerns. The supporting rationale or observable result is: Governance metadata explains retirement decisions and prevents unsafe unreviewed reactivation of an old artifact. This question describes a different issue: A training team creates a better model but reuses the prior version number in deployment notes. Address it through Assign and reference the distinct new registered model version in deployment records, with the required justification or evidence: Version references keep the deployed artifact unambiguous and preserve the history of successive model promotions.

B:

Record immutable review outcomes and approval timestamps with the model release metadata. This is useful for the distinct issue: An organization wants approval history for model versions, but reviewers communicate decisions only in private chat. The relevant technical rationale or evidence for that issue is: Audit trails should attach formal decisions to exact versions rather than relying on scattered conversations. Here the reported issue is: A training team creates a better model but reuses the prior version number in deployment notes. The relevant response is: Assign and reference the distinct new registered model version in deployment records. The first result cannot verify this different correction: Version references keep the deployed artifact unambiguous and preserve the history of successive model promotions.

C:

Retain or document the feature definitions and dependencies required to reproduce the archived model. It addresses another case: A model owner deletes a feature-retrieval dependency while archiving an earlier model version that may be needed for investigations. For that case, the distinguishing justification or acceptance check is: Weights without historical feature specifications may not be sufficient to reconstruct a model's past predictions. The present case instead states: A training team creates a better model but reuses the prior version number in deployment notes. It calls for Assign and reference the distinct new registered model version in deployment records. Showing success for the other case would not establish: Version references keep the deployed artifact unambiguous and preserve the history of successive model promotions.

Question 302

A team needs to show why an older model version stopped serving due to fairness concerns. What metadata should it keep?

  1. Inventory active deployment dependencies before retiring or deleting model versions.
  2. Promote the immutable historical version through a new documented deployment action.
  3. Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state.
  4. Point the production release reference to the approved active version explicitly.

Explanation

Correct Answer

C:

Governance metadata explains retirement decisions and prevents unsafe unreviewed reactivation of an old artifact.

Incorrect Answers

A:

Inventory active deployment dependencies before retiring or deleting model versions. That operation would be relevant if: A registry contains an old model still used by one critical production endpoint. An engineer proposes deleting every deprecated version. Its corresponding reason or verification is: A model version must not be removed while serving workloads still depend on its artifact and lineage. However the problem in this question is: A team needs to show why an older model version stopped serving due to fairness concerns. The requirement instead calls for Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state; the question-specific check is: Governance metadata explains retirement decisions and prevents unsafe unreviewed reactivation of an old artifact.

B:

Promote the immutable historical version through a new documented deployment action. A separate source case supports that approach when: An operator tries to roll back by editing an older model artifact in place rather than deploying its recorded version. The supporting rationale or observable result is: Rollback should restore a known artifact while preserving both the original failure and corrective deployment history. This question describes a different issue: A team needs to show why an older model version stopped serving due to fairness concerns. Address it through Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state, with the required justification or evidence: Governance metadata explains retirement decisions and prevents unsafe unreviewed reactivation of an old artifact.

D:

Point the production release reference to the approved active version explicitly. This is useful for the distinct issue: An archived model remains the default in a serving pipeline because its alias was never updated. The relevant technical rationale or evidence for that issue is: Archiving metadata alone does not guarantee every consumer stops resolving a deprecated model version. Here the reported issue is: A team needs to show why an older model version stopped serving due to fairness concerns. The relevant response is: Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state. The first result cannot verify this different correction: Governance metadata explains retirement decisions and prevents unsafe unreviewed reactivation of an old artifact.

Question 303

An archived model remains the default in a serving pipeline because its alias was never updated. What should be corrected?

  1. Retain an approved previous model version for controlled rollback.
  2. Point the production release reference to the approved active version explicitly.
  3. Maintain accountable owner and lifecycle status metadata for every registered model family.
  4. Use a stable registered model name and explicit version rather than transient storage paths.

Explanation

Correct Answer

B:

Archiving metadata alone does not guarantee every consumer stops resolving a deprecated model version.

Incorrect Answers

A:

The alternative, Retain an approved previous model version for controlled rollback, belongs to a different technical condition: A new candidate passes testing, but operators want a way to restore last week's production model quickly. There the source-specific mechanism or proof is: Preserving prior deployable artifacts supports recovery when a new model fails under real traffic. The condition asked about here is: An archived model remains the default in a serving pipeline because its alias was never updated. A suitable correction is Point the production release reference to the approved active version explicitly; the targeted justification or result is: Archiving metadata alone does not guarantee every consumer stops resolving a deprecated model version.

C:

Maintain accountable owner and lifecycle status metadata for every registered model family. That operation would be relevant if: A data science team has no record of who owns models abandoned in a shared registry. Its corresponding reason or verification is: Ownership information enables safe review, maintenance and retirement decisions when model-producing staff or projects change. However the problem in this question is: An archived model remains the default in a serving pipeline because its alias was never updated. The requirement instead calls for Point the production release reference to the approved active version explicitly; the question-specific check is: Archiving metadata alone does not guarantee every consumer stops resolving a deprecated model version.

D:

Use a stable registered model name and explicit version rather than transient storage paths. A separate source case supports that approach when: A production rollout references a model by its environment-specific storage URI, which changes after a workspace migration. The supporting rationale or observable result is: Registry identifiers separate logical model identity from storage location, improving portability across controlled deployments. This question describes a different issue: An archived model remains the default in a serving pipeline because its alias was never updated. Address it through Point the production release reference to the approved active version explicitly, with the required justification or evidence: Archiving metadata alone does not guarantee every consumer stops resolving a deprecated model version.

Question 304

A production rollout references a model by its environment-specific storage URI, which changes after a workspace migration. What more durable identifier should be used?

  1. Use a stable registered model name and explicit version rather than transient storage paths.
  2. Archive or remove obsolete model versions only after dependency and retention checks.
  3. Promote the immutable historical version through a new documented deployment action.
  4. Inventory active deployment dependencies before retiring or deleting model versions.

Explanation

Correct Answer

A:

Registry identifiers separate logical model identity from storage location, improving portability across controlled deployments.

Incorrect Answers

B:

Archive or remove obsolete model versions only after dependency and retention checks. It addresses another case: A decommissioned application no longer uses a production model, and storage costs grow with obsolete artifacts. For that case, the distinguishing justification or acceptance check is: Safe lifecycle cleanup balances operational cost with rollback needs and legally required artifact retention. The present case instead states: A production rollout references a model by its environment-specific storage URI, which changes after a workspace migration. It calls for Use a stable registered model name and explicit version rather than transient storage paths. Showing success for the other case would not establish: Registry identifiers separate logical model identity from storage location, improving portability across controlled deployments.

C:

The alternative, Promote the immutable historical version through a new documented deployment action, belongs to a different technical condition: An operator tries to roll back by editing an older model artifact in place rather than deploying its recorded version. There the source-specific mechanism or proof is: Rollback should restore a known artifact while preserving both the original failure and corrective deployment history. The condition asked about here is: A production rollout references a model by its environment-specific storage URI, which changes after a workspace migration. A suitable correction is Use a stable registered model name and explicit version rather than transient storage paths; the targeted justification or result is: Registry identifiers separate logical model identity from storage location, improving portability across controlled deployments.

D:

Inventory active deployment dependencies before retiring or deleting model versions. That operation would be relevant if: A registry contains an old model still used by one critical production endpoint. An engineer proposes deleting every deprecated version. Its corresponding reason or verification is: A model version must not be removed while serving workloads still depend on its artifact and lineage. However the problem in this question is: A production rollout references a model by its environment-specific storage URI, which changes after a workspace migration. The requirement instead calls for Use a stable registered model name and explicit version rather than transient storage paths; the question-specific check is: Registry identifiers separate logical model identity from storage location, improving portability across controlled deployments.

Question 305

An organization wants approval history for model versions, but reviewers communicate decisions only in private chat. What should the workflow create?

  1. Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state.
  2. Point the production release reference to the approved active version explicitly.
  3. Retain or document the feature definitions and dependencies required to reproduce the archived model.
  4. Record immutable review outcomes and approval timestamps with the model release metadata.

Explanation

Correct Answer

D:

Audit trails should attach formal decisions to exact versions rather than relying on scattered conversations.

Incorrect Answers

A:

Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state. This is useful for the distinct issue: A team needs to show why an older model version stopped serving due to fairness concerns. The relevant technical rationale or evidence for that issue is: Governance metadata explains retirement decisions and prevents unsafe unreviewed reactivation of an old artifact. Here the reported issue is: An organization wants approval history for model versions, but reviewers communicate decisions only in private chat. The relevant response is: Record immutable review outcomes and approval timestamps with the model release metadata. The first result cannot verify this different correction: Audit trails should attach formal decisions to exact versions rather than relying on scattered conversations.

B:

Point the production release reference to the approved active version explicitly. It addresses another case: An archived model remains the default in a serving pipeline because its alias was never updated. For that case, the distinguishing justification or acceptance check is: Archiving metadata alone does not guarantee every consumer stops resolving a deprecated model version. The present case instead states: An organization wants approval history for model versions, but reviewers communicate decisions only in private chat. It calls for Record immutable review outcomes and approval timestamps with the model release metadata. Showing success for the other case would not establish: Audit trails should attach formal decisions to exact versions rather than relying on scattered conversations.

C:

The alternative, Retain or document the feature definitions and dependencies required to reproduce the archived model, belongs to a different technical condition: A model owner deletes a feature-retrieval dependency while archiving an earlier model version that may be needed for investigations. There the source-specific mechanism or proof is: Weights without historical feature specifications may not be sufficient to reconstruct a model's past predictions. The condition asked about here is: An organization wants approval history for model versions, but reviewers communicate decisions only in private chat. A suitable correction is Record immutable review outcomes and approval timestamps with the model release metadata; the targeted justification or result is: Audit trails should attach formal decisions to exact versions rather than relying on scattered conversations.

Question 306

A model owner deletes a feature-retrieval dependency while archiving an earlier model version that may be needed for investigations. What should the retention plan include?

  1. Point the production release reference to the approved active version explicitly.
  2. Archive or remove obsolete model versions only after dependency and retention checks.
  3. Retain or document the feature definitions and dependencies required to reproduce the archived model.
  4. Promote the immutable historical version through a new documented deployment action.

Explanation

Correct Answer

C:

Weights without historical feature specifications may not be sufficient to reconstruct a model's past predictions.

Incorrect Answers

A:

Point the production release reference to the approved active version explicitly. A separate source case supports that approach when: An archived model remains the default in a serving pipeline because its alias was never updated. The supporting rationale or observable result is: Archiving metadata alone does not guarantee every consumer stops resolving a deprecated model version. This question describes a different issue: A model owner deletes a feature-retrieval dependency while archiving an earlier model version that may be needed for investigations. Address it through Retain or document the feature definitions and dependencies required to reproduce the archived model, with the required justification or evidence: Weights without historical feature specifications may not be sufficient to reconstruct a model's past predictions.

B:

Archive or remove obsolete model versions only after dependency and retention checks. This is useful for the distinct issue: A decommissioned application no longer uses a production model, and storage costs grow with obsolete artifacts. The relevant technical rationale or evidence for that issue is: Safe lifecycle cleanup balances operational cost with rollback needs and legally required artifact retention. Here the reported issue is: A model owner deletes a feature-retrieval dependency while archiving an earlier model version that may be needed for investigations. The relevant response is: Retain or document the feature definitions and dependencies required to reproduce the archived model. The first result cannot verify this different correction: Weights without historical feature specifications may not be sufficient to reconstruct a model's past predictions.

D:

Promote the immutable historical version through a new documented deployment action. It addresses another case: An operator tries to roll back by editing an older model artifact in place rather than deploying its recorded version. For that case, the distinguishing justification or acceptance check is: Rollback should restore a known artifact while preserving both the original failure and corrective deployment history. The present case instead states: A model owner deletes a feature-retrieval dependency while archiving an earlier model version that may be needed for investigations. It calls for Retain or document the feature definitions and dependencies required to reproduce the archived model. Showing success for the other case would not establish: Weights without historical feature specifications may not be sufficient to reconstruct a model's past predictions.

Question 307

An operator tries to roll back by editing an older model artifact in place rather than deploying its recorded version. What practice avoids provenance loss?

  1. Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state.
  2. Promote the immutable historical version through a new documented deployment action.
  3. Retain or document the feature definitions and dependencies required to reproduce the archived model.
  4. Archive or remove obsolete model versions only after dependency and retention checks.

Explanation

Correct Answer

B:

Rollback should restore a known artifact while preserving both the original failure and corrective deployment history.

Incorrect Answers

A:

Record the deprecation reason, reviewer decision, and replacement version alongside lifecycle state. That operation would be relevant if: A team needs to show why an older model version stopped serving due to fairness concerns. Its corresponding reason or verification is: Governance metadata explains retirement decisions and prevents unsafe unreviewed reactivation of an old artifact. However the problem in this question is: An operator tries to roll back by editing an older model artifact in place rather than deploying its recorded version. The requirement instead calls for Promote the immutable historical version through a new documented deployment action; the question-specific check is: Rollback should restore a known artifact while preserving both the original failure and corrective deployment history.

C:

Retain or document the feature definitions and dependencies required to reproduce the archived model. A separate source case supports that approach when: A model owner deletes a feature-retrieval dependency while archiving an earlier model version that may be needed for investigations. The supporting rationale or observable result is: Weights without historical feature specifications may not be sufficient to reconstruct a model's past predictions. This question describes a different issue: An operator tries to roll back by editing an older model artifact in place rather than deploying its recorded version. Address it through Promote the immutable historical version through a new documented deployment action, with the required justification or evidence: Rollback should restore a known artifact while preserving both the original failure and corrective deployment history.

D:

Archive or remove obsolete model versions only after dependency and retention checks. This is useful for the distinct issue: A decommissioned application no longer uses a production model, and storage costs grow with obsolete artifacts. The relevant technical rationale or evidence for that issue is: Safe lifecycle cleanup balances operational cost with rollback needs and legally required artifact retention. Here the reported issue is: An operator tries to roll back by editing an older model artifact in place rather than deploying its recorded version. The relevant response is: Promote the immutable historical version through a new documented deployment action. The first result cannot verify this different correction: Rollback should restore a known artifact while preserving both the original failure and corrective deployment history.

Question 308

A data science team has no record of who owns models abandoned in a shared registry. What lifecycle field is important?

  1. Maintain accountable owner and lifecycle status metadata for every registered model family.
  2. Inventory active deployment dependencies before retiring or deleting model versions.
  3. Retain the producing run, data versions, and code/environment lineage with the archived model.
  4. Point the production release reference to the approved active version explicitly.

Explanation

Correct Answer

A:

Ownership information enables safe review, maintenance and retirement decisions when model-producing staff or projects change.

Incorrect Answers

B:

The alternative, Inventory active deployment dependencies before retiring or deleting model versions, belongs to a different technical condition: A registry contains an old model still used by one critical production endpoint. An engineer proposes deleting every deprecated version. There the source-specific mechanism or proof is: A model version must not be removed while serving workloads still depend on its artifact and lineage. The condition asked about here is: A data science team has no record of who owns models abandoned in a shared registry. A suitable correction is Maintain accountable owner and lifecycle status metadata for every registered model family; the targeted justification or result is: Ownership information enables safe review, maintenance and retirement decisions when model-producing staff or projects change.

C:

Retain the producing run, data versions, and code/environment lineage with the archived model. That operation would be relevant if: A compliance audit asks which dataset and code produced a model version now retired from serving. Its corresponding reason or verification is: Archiving should preserve the evidence needed to reconstruct the model's origin despite removing it from active serving. However the problem in this question is: A data science team has no record of who owns models abandoned in a shared registry. The requirement instead calls for Maintain accountable owner and lifecycle status metadata for every registered model family; the question-specific check is: Ownership information enables safe review, maintenance and retirement decisions when model-producing staff or projects change.

D:

Point the production release reference to the approved active version explicitly. A separate source case supports that approach when: An archived model remains the default in a serving pipeline because its alias was never updated. The supporting rationale or observable result is: Archiving metadata alone does not guarantee every consumer stops resolving a deprecated model version. This question describes a different issue: A data science team has no record of who owns models abandoned in a shared registry. Address it through Maintain accountable owner and lifecycle status metadata for every registered model family, with the required justification or evidence: Ownership information enables safe review, maintenance and retirement decisions when model-producing staff or projects change.

Question 309

An ecommerce fraud check must return a prediction synchronously within 100 milliseconds of a payment request. What Azure ML serving interface fits?

  1. Use a supported bring-your-own-container managed online deployment.
  2. Increase appropriate deployment instance capacity or configure supported autoscaling.
  3. Use an Azure Machine Learning batch endpoint or batch inference job.
  4. Deploy the model to a managed online endpoint for real-time inference.

Explanation

Correct Answer

D:

Online endpoints accept synchronous requests, unlike offline batch jobs that process larger workloads asynchronously.

Incorrect Answers

A:

Use a supported bring-your-own-container managed online deployment. It addresses another case: A model requires a proprietary inference runtime packed as a complete image, not a conventional Python scoring script. For that case, the distinguishing justification or acceptance check is: Custom containers can package specialized runtimes while still using managed endpoint lifecycle and network features. The present case instead states: An ecommerce fraud check must return a prediction synchronously within 100 milliseconds of a payment request. It calls for Deploy the model to a managed online endpoint for real-time inference. Showing success for the other case would not establish: Online endpoints accept synchronous requests, unlike offline batch jobs that process larger workloads asynchronously.

B:

The alternative, Increase appropriate deployment instance capacity or configure supported autoscaling, belongs to a different technical condition: An endpoint experiences high CPU utilization at peak traffic and the team wants capacity to handle simultaneous calls. There the source-specific mechanism or proof is: Online serving throughput depends on deployed compute and replica counts, not only the model registry version. The condition asked about here is: An ecommerce fraud check must return a prediction synchronously within 100 milliseconds of a payment request. A suitable correction is Deploy the model to a managed online endpoint for real-time inference; the targeted justification or result is: Online endpoints accept synchronous requests, unlike offline batch jobs that process larger workloads asynchronously.

C:

Use an Azure Machine Learning batch endpoint or batch inference job. That operation would be relevant if: A monthly reporting job scores 30 million records overnight without per-row API latency requirements. Its corresponding reason or verification is: Batch endpoints are designed for asynchronous bulk scoring rather than a continually invoked low-latency API. However the problem in this question is: An ecommerce fraud check must return a prediction synchronously within 100 milliseconds of a payment request. The requirement instead calls for Deploy the model to a managed online endpoint for real-time inference; the question-specific check is: Online endpoints accept synchronous requests, unlike offline batch jobs that process larger workloads asynchronously.

Question 310

A monthly reporting job scores 30 million records overnight without per-row API latency requirements. What serving option is appropriate?

  1. Use online inference for synchronous responses or treat batch endpoint calls as asynchronous jobs.
  2. Configure supported private network access and authentication for the managed serving endpoint.
  3. Use an Azure Machine Learning batch endpoint or batch inference job.
  4. Move credentials to a protected secret service or use identity-based access where supported.

Explanation

Correct Answer

C:

Batch endpoints are designed for asynchronous bulk scoring rather than a continually invoked low-latency API.

Incorrect Answers

A:

Use online inference for synchronous responses or treat batch endpoint calls as asynchronous jobs. It addresses another case: A batch endpoint processes files from a protected data store but users expect an immediate JSON prediction response for each file. For that case, the distinguishing justification or acceptance check is: Batch endpoints execute workloads asynchronously and return job status, not interactive predictions in the immediate request response. The present case instead states: A monthly reporting job scores 30 million records overnight without per-row API latency requirements. It calls for Use an Azure Machine Learning batch endpoint or batch inference job. Showing success for the other case would not establish: Batch endpoints are designed for asynchronous bulk scoring rather than a continually invoked low-latency API.

B:

The alternative, Configure supported private network access and authentication for the managed serving endpoint, belongs to a different technical condition: A production system needs to keep inference traffic on approved private networks and avoid public access. There the source-specific mechanism or proof is: Serving interface exposure is independent from model quality; network controls must enforce the application's required perimeter. The condition asked about here is: A monthly reporting job scores 30 million records overnight without per-row API latency requirements. A suitable correction is Use an Azure Machine Learning batch endpoint or batch inference job; the targeted justification or result is: Batch endpoints are designed for asynchronous bulk scoring rather than a continually invoked low-latency API.

D:

Move credentials to a protected secret service or use identity-based access where supported. That operation would be relevant if: A public application uses a single fixed model-serving key stored in source code. Its corresponding reason or verification is: Secrets in application source are easily copied and difficult to rotate independently of software releases. However the problem in this question is: A monthly reporting job scores 30 million records overnight without per-row API latency requirements. The requirement instead calls for Use an Azure Machine Learning batch endpoint or batch inference job; the question-specific check is: Batch endpoints are designed for asynchronous bulk scoring rather than a continually invoked low-latency API.

Operationalizing Machine Learning and Generative AI Solutions (AI-300) Practice Tests