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 2

Question 10

A workspace must be retired, but its model versions and audit artifacts are still needed for an open investigation. What should precede deletion?

  1. Identify the workspace's dependent storage and other linked services, then protect or replace required dependencies.
  2. Request the applicable regional VM-family quota or select available supported capacity.
  3. Inventory and export required model, run, and governance records into retained storage.
  4. Apply meaningful resource tags and scoped cost allocation to the departments' workspaces and compute.

Explanation

Correct Answer

C:

Deletion can remove the workspace's accessible artifacts; preserving evidence is a data-retention decision separate from stopping compute.

Incorrect Answers

A:

Identify the workspace's dependent storage and other linked services, then protect or replace required dependencies. This is useful for the distinct issue: A research workspace uses a shared storage account that a platform team will delete during cleanup. The relevant technical rationale or evidence for that issue is: The workspace depends on linked Azure resources for artifacts and operations; deleting storage without migration can break active experiments. Here the reported issue is: A workspace must be retired, but its model versions and audit artifacts are still needed for an open investigation. The relevant response is: Inventory and export required model, run, and governance records into retained storage. The first result cannot verify this different correction: Deletion can remove the workspace's accessible artifacts; preserving evidence is a data-retention decision separate from stopping compute.

B:

Request the applicable regional VM-family quota or select available supported capacity. It addresses another case: A workspace administrator cannot create new experiments because the requested region has insufficient VM quota. For that case, the distinguishing justification or acceptance check is: Compute quota is regional and distinct from workspace metadata; blindly recreating the workspace does not increase available compute quota. The present case instead states: A workspace must be retired, but its model versions and audit artifacts are still needed for an open investigation. It calls for Inventory and export required model, run, and governance records into retained storage. Showing success for the other case would not establish: Deletion can remove the workspace's accessible artifacts; preserving evidence is a data-retention decision separate from stopping compute.

D:

The alternative, Apply meaningful resource tags and scoped cost allocation to the departments' workspaces and compute, belongs to a different technical condition: Two departments need separate chargeback reports for Azure ML training compute. They currently run all jobs in one workspace with no cost attribution. There the source-specific mechanism or proof is: Resource organization and cost tags support accountable consumption tracking; changing training code alone does not attribute Azure spending. The condition asked about here is: A workspace must be retired, but its model versions and audit artifacts are still needed for an open investigation. A suitable correction is Inventory and export required model, run, and governance records into retained storage; the targeted justification or result is: Deletion can remove the workspace's accessible artifacts; preserving evidence is a data-retention decision separate from stopping compute.

Question 11

An MLOps team wants every new workspace to automatically connect to the correct Key Vault, Storage, and Application Insights dependencies. What should be configured?

  1. Set the workspace location to the approved European region before provisioning.
  2. Declare and verify the intended dependent Azure resources in the workspace deployment definition.
  3. Deploy the workspace configuration from version-controlled infrastructure-as-code templates.
  4. Use a stable workspace name and idempotent declared infrastructure configuration per environment.

Explanation

Correct Answer

B:

Explicit dependency mapping prevents an accidentally created workspace from relying on unintended shared services or missing platform dependencies.

Incorrect Answers

A:

Set the workspace location to the approved European region before provisioning. A separate source case supports that approach when: A regulated training team must keep its new Azure Machine Learning workspace in an approved European Azure region. A provisioning template defaults to East US. The supporting rationale or observable result is: The workspace resource location is selected at creation; tagging an East US deployment cannot relocate its services or data dependencies. This question describes a different issue: An MLOps team wants every new workspace to automatically connect to the correct Key Vault, Storage, and Application Insights dependencies. Address it through Declare and verify the intended dependent Azure resources in the workspace deployment definition, with the required justification or evidence: Explicit dependency mapping prevents an accidentally created workspace from relying on unintended shared services or missing platform dependencies.

C:

Deploy the workspace configuration from version-controlled infrastructure-as-code templates. This is useful for the distinct issue: Several workspaces require a centrally consistent policy for encryption, tagging, and access. Manual workspace creation keeps drifting. The relevant technical rationale or evidence for that issue is: Declared resource configuration enables peer review, consistency, and reproducible environments across teams instead of untracked portal changes. Here the reported issue is: An MLOps team wants every new workspace to automatically connect to the correct Key Vault, Storage, and Application Insights dependencies. The relevant response is: Declare and verify the intended dependent Azure resources in the workspace deployment definition. The first result cannot verify this different correction: Explicit dependency mapping prevents an accidentally created workspace from relying on unintended shared services or missing platform dependencies.

D:

Use a stable workspace name and idempotent declared infrastructure configuration per environment. It addresses another case: A corporate script repeatedly creates duplicate Azure ML workspaces because each execution uses a random resource name. For that case, the distinguishing justification or acceptance check is: Consistent resource identifiers allow repeat deployments to update the intended resource instead of proliferating unmanaged workspaces. The present case instead states: An MLOps team wants every new workspace to automatically connect to the correct Key Vault, Storage, and Application Insights dependencies. It calls for Declare and verify the intended dependent Azure resources in the workspace deployment definition. Showing success for the other case would not establish: Explicit dependency mapping prevents an accidentally created workspace from relying on unintended shared services or missing platform dependencies.

Question 12

Azure ML training jobs must read a protected blob container without embedding a storage account key in code. What access approach is appropriate?

  1. Create a named datastore targeting the governed production account and use it in job inputs.
  2. Renew the expired delegated credential or migrate the connection to managed identity access.
  3. Configure an identity-based datastore connection and grant least-privilege storage data access.
  4. Grant storage data access to the job's runtime identity and verify datastore authentication mode.

Explanation

Correct Answer

C:

Managed identity and Azure RBAC can authorize storage reads without distributing persistent account keys through training scripts.

Incorrect Answers

A:

Create a named datastore targeting the governed production account and use it in job inputs. That operation would be relevant if: A pipeline reads from its workspace default datastore, but production data belongs in a separate governed storage account. Its corresponding reason or verification is: Explicit data locations avoid silently relying on the default backing store and make production access auditable. However the problem in this question is: Azure ML training jobs must read a protected blob container without embedding a storage account key in code. The requirement instead calls for Configure an identity-based datastore connection and grant least-privilege storage data access; the question-specific check is: Managed identity and Azure RBAC can authorize storage reads without distributing persistent account keys through training scripts.

B:

Renew the expired delegated credential or migrate the connection to managed identity access. A separate source case supports that approach when: A datastore configured with a temporary SAS token worked yesterday but now returns 403 responses on every job. The supporting rationale or observable result is: Time-limited SAS credentials expire independently of the model code; credential renewal or identity-based access restores authorization. This question describes a different issue: Azure ML training jobs must read a protected blob container without embedding a storage account key in code. Address it through Configure an identity-based datastore connection and grant least-privilege storage data access, with the required justification or evidence: Managed identity and Azure RBAC can authorize storage reads without distributing persistent account keys through training scripts.

D:

Grant storage data access to the job's runtime identity and verify datastore authentication mode. This is useful for the distinct issue: A job can read a datastore from a user's notebook but fails on a managed compute cluster. The relevant technical rationale or evidence for that issue is: Interactive user authorization may not transfer to managed job identities; compute execution must possess its own permitted access. Here the reported issue is: Azure ML training jobs must read a protected blob container without embedding a storage account key in code. The relevant response is: Configure an identity-based datastore connection and grant least-privilege storage data access. The first result cannot verify this different correction: Managed identity and Azure RBAC can authorize storage reads without distributing persistent account keys through training scripts.

Question 13

An Azure ML job expects nested directories in an ADLS Gen2 container, but its configured datastore points to a different blob root. What should the team correct?

  1. Confirm data-plane permissions and private network connectivity without opening public access.
  2. Configure the datastore to reference the intended ADLS Gen2 filesystem and directory path.
  3. Create a named datastore targeting the governed production account and use it in job inputs.
  4. Register authorized datastores that reference the same permitted storage location.

Explanation

Correct Answer

B:

The datastore must identify the actual backing storage location; changing experiment names cannot fix a misaddressed filesystem.

Incorrect Answers

A:

The alternative, Confirm data-plane permissions and private network connectivity without opening public access, belongs to a different technical condition: An engineer proposes making training storage public to resolve datastore read failures. There the source-specific mechanism or proof is: Public exposure is unnecessary when the real fault may be identity or private endpoint routing configuration. The condition asked about here is: An Azure ML job expects nested directories in an ADLS Gen2 container, but its configured datastore points to a different blob root. A suitable correction is Configure the datastore to reference the intended ADLS Gen2 filesystem and directory path; the targeted justification or result is: The datastore must identify the actual backing storage location; changing experiment names cannot fix a misaddressed filesystem.

C:

Create a named datastore targeting the governed production account and use it in job inputs. That operation would be relevant if: A pipeline reads from its workspace default datastore, but production data belongs in a separate governed storage account. Its corresponding reason or verification is: Explicit data locations avoid silently relying on the default backing store and make production access auditable. However the problem in this question is: An Azure ML job expects nested directories in an ADLS Gen2 container, but its configured datastore points to a different blob root. The requirement instead calls for Configure the datastore to reference the intended ADLS Gen2 filesystem and directory path; the question-specific check is: The datastore must identify the actual backing storage location; changing experiment names cannot fix a misaddressed filesystem.

D:

Register authorized datastores that reference the same permitted storage location. A separate source case supports that approach when: Two Azure ML training teams need read-only access to the same large files without copying them into each workspace. The supporting rationale or observable result is: Datastores provide reusable pointers and access configuration without forcing another physical copy of every dataset. This question describes a different issue: An Azure ML job expects nested directories in an ADLS Gen2 container, but its configured datastore points to a different blob root. Address it through Configure the datastore to reference the intended ADLS Gen2 filesystem and directory path, with the required justification or evidence: The datastore must identify the actual backing storage location; changing experiment names cannot fix a misaddressed filesystem.

Question 14

A pipeline reads from its workspace default datastore, but production data belongs in a separate governed storage account. Which change makes data access explicit?

  1. Create a named datastore targeting the governed production account and use it in job inputs.
  2. Register authorized datastores that reference the same permitted storage location.
  3. Update the datastore connection's credential reference and retest read permissions.
  4. Set output paths to an approved writable datastore rather than the reference data location.

Explanation

Correct Answer

A:

Explicit data locations avoid silently relying on the default backing store and make production access auditable.

Incorrect Answers

B:

Register authorized datastores that reference the same permitted storage location. It addresses another case: Two Azure ML training teams need read-only access to the same large files without copying them into each workspace. For that case, the distinguishing justification or acceptance check is: Datastores provide reusable pointers and access configuration without forcing another physical copy of every dataset. The present case instead states: A pipeline reads from its workspace default datastore, but production data belongs in a separate governed storage account. It calls for Create a named datastore targeting the governed production account and use it in job inputs. Showing success for the other case would not establish: Explicit data locations avoid silently relying on the default backing store and make production access auditable.

C:

The alternative, Update the datastore connection's credential reference and retest read permissions, belongs to a different technical condition: An MLOps engineer rotates the credential used by a non-identity datastore and all future jobs fail. There the source-specific mechanism or proof is: Credential rotation requires dependent connections to use the new credential; editing an unrelated asset version cannot repair authentication. The condition asked about here is: A pipeline reads from its workspace default datastore, but production data belongs in a separate governed storage account. A suitable correction is Create a named datastore targeting the governed production account and use it in job inputs; the targeted justification or result is: Explicit data locations avoid silently relying on the default backing store and make production access auditable.

D:

Set output paths to an approved writable datastore rather than the reference data location. That operation would be relevant if: Developers accidentally write training outputs into a read-only reference datastore. Its corresponding reason or verification is: Input and output storage permissions can differ; a writable output destination is needed for pipeline artifacts. However the problem in this question is: A pipeline reads from its workspace default datastore, but production data belongs in a separate governed storage account. The requirement instead calls for Create a named datastore targeting the governed production account and use it in job inputs; the question-specific check is: Explicit data locations avoid silently relying on the default backing store and make production access auditable.

Question 15

A datastore configured with a temporary SAS token worked yesterday but now returns 403 responses on every job. What is the likely remediation?

  1. Confirm data-plane permissions and private network connectivity without opening public access.
  2. Configure the datastore to reference the intended ADLS Gen2 filesystem and directory path.
  3. Review datastore resource references and cross-subscription identity permissions explicitly.
  4. Renew the expired delegated credential or migrate the connection to managed identity access.

Explanation

Correct Answer

D:

Time-limited SAS credentials expire independently of the model code; credential renewal or identity-based access restores authorization.

Incorrect Answers

A:

Confirm data-plane permissions and private network connectivity without opening public access. This is useful for the distinct issue: An engineer proposes making training storage public to resolve datastore read failures. The relevant technical rationale or evidence for that issue is: Public exposure is unnecessary when the real fault may be identity or private endpoint routing configuration. Here the reported issue is: A datastore configured with a temporary SAS token worked yesterday but now returns 403 responses on every job. The relevant response is: Renew the expired delegated credential or migrate the connection to managed identity access. The first result cannot verify this different correction: Time-limited SAS credentials expire independently of the model code; credential renewal or identity-based access restores authorization.

B:

Configure the datastore to reference the intended ADLS Gen2 filesystem and directory path. It addresses another case: An Azure ML job expects nested directories in an ADLS Gen2 container, but its configured datastore points to a different blob root. For that case, the distinguishing justification or acceptance check is: The datastore must identify the actual backing storage location; changing experiment names cannot fix a misaddressed filesystem. The present case instead states: A datastore configured with a temporary SAS token worked yesterday but now returns 403 responses on every job. It calls for Renew the expired delegated credential or migrate the connection to managed identity access. Showing success for the other case would not establish: Time-limited SAS credentials expire independently of the model code; credential renewal or identity-based access restores authorization.

C:

The alternative, Review datastore resource references and cross-subscription identity permissions explicitly, belongs to a different technical condition: A workspace is moved to a new subscription and its datastore still points to the previous subscription's storage account. There the source-specific mechanism or proof is: Subscription changes do not transparently update storage references or their RBAC; the new workspace needs valid data-plane authorization. The condition asked about here is: A datastore configured with a temporary SAS token worked yesterday but now returns 403 responses on every job. A suitable correction is Renew the expired delegated credential or migrate the connection to managed identity access; the targeted justification or result is: Time-limited SAS credentials expire independently of the model code; credential renewal or identity-based access restores authorization.

Question 16

Two Azure ML training teams need read-only access to the same large files without copying them into each workspace. What is the efficient configuration?

  1. Renew the expired delegated credential or migrate the connection to managed identity access.
  2. Review datastore resource references and cross-subscription identity permissions explicitly.
  3. Register authorized datastores that reference the same permitted storage location.
  4. Confirm data-plane permissions and private network connectivity without opening public access.

Explanation

Correct Answer

C:

Datastores provide reusable pointers and access configuration without forcing another physical copy of every dataset.

Incorrect Answers

A:

Renew the expired delegated credential or migrate the connection to managed identity access. A separate source case supports that approach when: A datastore configured with a temporary SAS token worked yesterday but now returns 403 responses on every job. The supporting rationale or observable result is: Time-limited SAS credentials expire independently of the model code; credential renewal or identity-based access restores authorization. This question describes a different issue: Two Azure ML training teams need read-only access to the same large files without copying them into each workspace. Address it through Register authorized datastores that reference the same permitted storage location, with the required justification or evidence: Datastores provide reusable pointers and access configuration without forcing another physical copy of every dataset.

B:

Review datastore resource references and cross-subscription identity permissions explicitly. This is useful for the distinct issue: A workspace is moved to a new subscription and its datastore still points to the previous subscription's storage account. The relevant technical rationale or evidence for that issue is: Subscription changes do not transparently update storage references or their RBAC; the new workspace needs valid data-plane authorization. Here the reported issue is: Two Azure ML training teams need read-only access to the same large files without copying them into each workspace. The relevant response is: Register authorized datastores that reference the same permitted storage location. The first result cannot verify this different correction: Datastores provide reusable pointers and access configuration without forcing another physical copy of every dataset.

D:

Confirm data-plane permissions and private network connectivity without opening public access. It addresses another case: An engineer proposes making training storage public to resolve datastore read failures. For that case, the distinguishing justification or acceptance check is: Public exposure is unnecessary when the real fault may be identity or private endpoint routing configuration. The present case instead states: Two Azure ML training teams need read-only access to the same large files without copying them into each workspace. It calls for Register authorized datastores that reference the same permitted storage location. Showing success for the other case would not establish: Datastores provide reusable pointers and access configuration without forcing another physical copy of every dataset.

Question 17

A datastore shows healthy metadata but read operations fail only after Storage firewall restrictions were enabled. What should be investigated?

  1. Update the datastore connection's credential reference and retest read permissions.
  2. Check workspace or compute network path and storage firewall/private endpoint rules for the datastore.
  3. Create a named datastore targeting the governed production account and use it in job inputs.
  4. Review datastore resource references and cross-subscription identity permissions explicitly.

Explanation

Correct Answer

B:

A valid datastore reference does not override network denial; the data plane must be reachable from executing compute.

Incorrect Answers

A:

Update the datastore connection's credential reference and retest read permissions. That operation would be relevant if: An MLOps engineer rotates the credential used by a non-identity datastore and all future jobs fail. Its corresponding reason or verification is: Credential rotation requires dependent connections to use the new credential; editing an unrelated asset version cannot repair authentication. However the problem in this question is: A datastore shows healthy metadata but read operations fail only after Storage firewall restrictions were enabled. The requirement instead calls for Check workspace or compute network path and storage firewall/private endpoint rules for the datastore; the question-specific check is: A valid datastore reference does not override network denial; the data plane must be reachable from executing compute.

C:

Create a named datastore targeting the governed production account and use it in job inputs. A separate source case supports that approach when: A pipeline reads from its workspace default datastore, but production data belongs in a separate governed storage account. The supporting rationale or observable result is: Explicit data locations avoid silently relying on the default backing store and make production access auditable. This question describes a different issue: A datastore shows healthy metadata but read operations fail only after Storage firewall restrictions were enabled. Address it through Check workspace or compute network path and storage firewall/private endpoint rules for the datastore, with the required justification or evidence: A valid datastore reference does not override network denial; the data plane must be reachable from executing compute.

D:

Review datastore resource references and cross-subscription identity permissions explicitly. This is useful for the distinct issue: A workspace is moved to a new subscription and its datastore still points to the previous subscription's storage account. The relevant technical rationale or evidence for that issue is: Subscription changes do not transparently update storage references or their RBAC; the new workspace needs valid data-plane authorization. Here the reported issue is: A datastore shows healthy metadata but read operations fail only after Storage firewall restrictions were enabled. The relevant response is: Check workspace or compute network path and storage firewall/private endpoint rules for the datastore. The first result cannot verify this different correction: A valid datastore reference does not override network denial; the data plane must be reachable from executing compute.

Question 18

An MLOps engineer rotates the credential used by a non-identity datastore and all future jobs fail. What operational change is necessary?

  1. Update the datastore connection's credential reference and retest read permissions.
  2. Renew the expired delegated credential or migrate the connection to managed identity access.
  3. Configure an identity-based datastore connection and grant least-privilege storage data access.
  4. Register authorized datastores that reference the same permitted storage location.

Explanation

Correct Answer

A:

Credential rotation requires dependent connections to use the new credential; editing an unrelated asset version cannot repair authentication.

Incorrect Answers

B:

The alternative, Renew the expired delegated credential or migrate the connection to managed identity access, belongs to a different technical condition: A datastore configured with a temporary SAS token worked yesterday but now returns 403 responses on every job. There the source-specific mechanism or proof is: Time-limited SAS credentials expire independently of the model code; credential renewal or identity-based access restores authorization. The condition asked about here is: An MLOps engineer rotates the credential used by a non-identity datastore and all future jobs fail. A suitable correction is Update the datastore connection's credential reference and retest read permissions; the targeted justification or result is: Credential rotation requires dependent connections to use the new credential; editing an unrelated asset version cannot repair authentication.

C:

Configure an identity-based datastore connection and grant least-privilege storage data access. That operation would be relevant if: Azure ML training jobs must read a protected blob container without embedding a storage account key in code. Its corresponding reason or verification is: Managed identity and Azure RBAC can authorize storage reads without distributing persistent account keys through training scripts. However the problem in this question is: An MLOps engineer rotates the credential used by a non-identity datastore and all future jobs fail. The requirement instead calls for Update the datastore connection's credential reference and retest read permissions; the question-specific check is: Credential rotation requires dependent connections to use the new credential; editing an unrelated asset version cannot repair authentication.

D:

Register authorized datastores that reference the same permitted storage location. A separate source case supports that approach when: Two Azure ML training teams need read-only access to the same large files without copying them into each workspace. The supporting rationale or observable result is: Datastores provide reusable pointers and access configuration without forcing another physical copy of every dataset. This question describes a different issue: An MLOps engineer rotates the credential used by a non-identity datastore and all future jobs fail. Address it through Update the datastore connection's credential reference and retest read permissions, with the required justification or evidence: Credential rotation requires dependent connections to use the new credential; editing an unrelated asset version cannot repair authentication.

Question 19

Developers accidentally write training outputs into a read-only reference datastore. How should the pipeline express the intended write target?

  1. Configure the datastore to reference the intended ADLS Gen2 filesystem and directory path.
  2. Configure an identity-based datastore connection and grant least-privilege storage data access.
  3. Create a named datastore targeting the governed production account and use it in job inputs.
  4. Set output paths to an approved writable datastore rather than the reference data location.

Explanation

Correct Answer

D:

Input and output storage permissions can differ; a writable output destination is needed for pipeline artifacts.

Incorrect Answers

A:

Configure the datastore to reference the intended ADLS Gen2 filesystem and directory path. It addresses another case: An Azure ML job expects nested directories in an ADLS Gen2 container, but its configured datastore points to a different blob root. For that case, the distinguishing justification or acceptance check is: The datastore must identify the actual backing storage location; changing experiment names cannot fix a misaddressed filesystem. The present case instead states: Developers accidentally write training outputs into a read-only reference datastore. It calls for Set output paths to an approved writable datastore rather than the reference data location. Showing success for the other case would not establish: Input and output storage permissions can differ; a writable output destination is needed for pipeline artifacts.

B:

The alternative, Configure an identity-based datastore connection and grant least-privilege storage data access, belongs to a different technical condition: Azure ML training jobs must read a protected blob container without embedding a storage account key in code. There the source-specific mechanism or proof is: Managed identity and Azure RBAC can authorize storage reads without distributing persistent account keys through training scripts. The condition asked about here is: Developers accidentally write training outputs into a read-only reference datastore. A suitable correction is Set output paths to an approved writable datastore rather than the reference data location; the targeted justification or result is: Input and output storage permissions can differ; a writable output destination is needed for pipeline artifacts.

C:

Create a named datastore targeting the governed production account and use it in job inputs. That operation would be relevant if: A pipeline reads from its workspace default datastore, but production data belongs in a separate governed storage account. Its corresponding reason or verification is: Explicit data locations avoid silently relying on the default backing store and make production access auditable. However the problem in this question is: Developers accidentally write training outputs into a read-only reference datastore. The requirement instead calls for Set output paths to an approved writable datastore rather than the reference data location; the question-specific check is: Input and output storage permissions can differ; a writable output destination is needed for pipeline artifacts.

Question 20

A job can read a datastore from a user's notebook but fails on a managed compute cluster. Why should the platform team inspect identity context?

  1. Configure the datastore to reference the intended ADLS Gen2 filesystem and directory path.
  2. Review datastore resource references and cross-subscription identity permissions explicitly.
  3. Grant storage data access to the job's runtime identity and verify datastore authentication mode.
  4. Renew the expired delegated credential or migrate the connection to managed identity access.

Explanation

Correct Answer

C:

Interactive user authorization may not transfer to managed job identities; compute execution must possess its own permitted access.

Incorrect Answers

A:

Configure the datastore to reference the intended ADLS Gen2 filesystem and directory path. It addresses another case: An Azure ML job expects nested directories in an ADLS Gen2 container, but its configured datastore points to a different blob root. For that case, the distinguishing justification or acceptance check is: The datastore must identify the actual backing storage location; changing experiment names cannot fix a misaddressed filesystem. The present case instead states: A job can read a datastore from a user's notebook but fails on a managed compute cluster. It calls for Grant storage data access to the job's runtime identity and verify datastore authentication mode. Showing success for the other case would not establish: Interactive user authorization may not transfer to managed job identities; compute execution must possess its own permitted access.

B:

The alternative, Review datastore resource references and cross-subscription identity permissions explicitly, belongs to a different technical condition: A workspace is moved to a new subscription and its datastore still points to the previous subscription's storage account. There the source-specific mechanism or proof is: Subscription changes do not transparently update storage references or their RBAC; the new workspace needs valid data-plane authorization. The condition asked about here is: A job can read a datastore from a user's notebook but fails on a managed compute cluster. A suitable correction is Grant storage data access to the job's runtime identity and verify datastore authentication mode; the targeted justification or result is: Interactive user authorization may not transfer to managed job identities; compute execution must possess its own permitted access.

D:

Renew the expired delegated credential or migrate the connection to managed identity access. That operation would be relevant if: A datastore configured with a temporary SAS token worked yesterday but now returns 403 responses on every job. Its corresponding reason or verification is: Time-limited SAS credentials expire independently of the model code; credential renewal or identity-based access restores authorization. However the problem in this question is: A job can read a datastore from a user's notebook but fails on a managed compute cluster. The requirement instead calls for Grant storage data access to the job's runtime identity and verify datastore authentication mode; the question-specific check is: Interactive user authorization may not transfer to managed job identities; compute execution must possess its own permitted access.

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