Pass Microsoft AI-103 Exam in First Attempt Guaranteed!
Get 100% Latest Exam Questions, Accurate & Verified Answers!
30 Days Free Updates, Instant Download!
Microsoft AI-103 Practice Test 78
Question 770
Production requirement: A procurement inbox needs supplier names and purchase-order IDs extracted from mixed-format emails. Which TWO decisions would undermine the required system boundary? Select TWO answers.
Explanation
Correct Answer
“Treat every capitalized word as an organization entity” would violate the requirement. The appropriate control instead is to define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review.
“Invent missing account numbers to satisfy a required field” would violate the requirement. The appropriate control instead is to define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review.
Incorrect Answers
“Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented” is the required outcome of a working control, not an action that weakens it. Confirm that extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
“Inspect records that explain the affected process, then check the result described in this case: A procurement inbox needs supplier names and purchase-order IDs extracted from mixed-format emails” adds a relevant safeguard or verification step rather than weakening the solution. Confirm that extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
“Define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review” is the controlled implementation, not an unsafe workaround.
Question 771
A team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. Which design decision is justified by the failure and the required outcome?
Explanation
Correct Answer
Because a team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities, entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Incorrect Answers
A team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. “Normalize multiple unrelated people to one identifier” does not implement the control that the facts call for. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Instead, define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review.
A team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. “Discard the source span that supports an entity label” does not implement the control that the facts call for. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Instead, define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review.
“Apply the proposed setting to all connected workloads before testing the case-specific boundary: a health insurer wants medication” does not implement the control that the facts call for. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Instead, define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review.
Question 772
Situation: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. Which underlying technical principle explains the recommended implementation?
Explanation
Correct Answer
Because a team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities, entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review. The decisive evidence is that extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Incorrect Answers
A team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. “Successful provisioning by itself proves all security, output, and operational constraints are satisfied” does not fully explain the decisive technical boundary in this case. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
“The fact that the team started work is enough to prove correctness for this exact case: a health insurer wants medication” does not fully explain the decisive technical boundary in this case. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
A team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. “The most powerful or least restricted model is always the safest choice regardless of task conditions” does not fully explain the decisive technical boundary in this case. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. The implementation must define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review.
Question 773
Architecture case: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. What observable acceptance condition should be required before closing the task?
Explanation
Correct Answer
A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Incorrect Answers
A team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. “The team closes a change ticket without replaying a representative positive and negative test” alone does not establish all the required behavior. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
A team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. “A project dashboard remains green, although the affected request, output and permissions were not checked” would not meet the requirement. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
“The operator reports success without reproducing the observed behavior: a health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities” alone does not establish all the required behavior. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
Question 774
Production requirement: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. Which first investigation is more useful than an unrelated health indicator?
Explanation
Correct Answer
Because a team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities, compare source offsets, label definitions, confidence or evidence, schema errors and human adjudication examples. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Incorrect Answers
“Read only the broad service-health summary and skip actual workload evidence” would not meet the requirement. Compare source offsets, label definitions, confidence or evidence, schema errors and human adjudication examples.
“Restart every affected component before collecting request or tool traces” would not be the most direct first check for this failure. Compare source offsets, label definitions, confidence or evidence, schema errors and human adjudication examples.
“Assume the general service is at fault without checking the affected operation in this case: a health insurer wants medication” would not meet the requirement. Compare source offsets, label definitions, confidence or evidence, schema errors and human adjudication examples.
Question 775
A team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. Which TWO actions could conceal a serious failure while appearing faster? Select TWO answers.
Explanation
Correct Answer
A team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. “Invent missing account numbers to satisfy a required field” would violate the requirement. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
A team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. “Treat every capitalized word as an organization entity” would violate the requirement. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
Incorrect Answers
A team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities. “Define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review” is the controlled implementation, not an unsafe workaround. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
“Inspect records that explain the affected process, then check the result described in this case: A health insurer wants medication” adds a relevant safeguard or verification step rather than weakening the solution. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Confirm that extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Because a team is investigating this task: A health insurer wants medication and dosage mentions captured with source spans instead of hallucinated missing quantities, “Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented” is the required outcome of a working control, not an action that weakens it. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Confirm that extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Question 776
Situation: A city licensing office receives abbreviations that map to different local agencies depending on context. Which implementation best meets the requirement while honoring the stated constraints?
Explanation
Correct Answer
A city licensing office receives abbreviations that map to different local agencies depending on context. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Incorrect Answers
Because a city licensing office receives abbreviations that map to different local agencies depending on context, “Discard the source span that supports an entity label” does not implement the control that the facts call for. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Instead, define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review.
“Apply the proposed setting to all connected workloads before testing the case-specific boundary: a city licensing office receives abbreviations that map to different local agencies depending on context” does not implement the control that the facts call for. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Instead, define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review.
“Normalize multiple unrelated people to one identifier” does not implement the control that the facts call for. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Instead, define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review.
Question 777
Architecture case: A city licensing office receives abbreviations that map to different local agencies depending on context. Which design boundary is most relevant to this decision?
Explanation
Correct Answer
A city licensing office receives abbreviations that map to different local agencies depending on context. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review. The decisive evidence is that extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Incorrect Answers
“The fact that the team started work is enough to prove correctness for this exact case: a city licensing office receives abbreviations that map to different local agencies depending on context” does not fully explain the decisive technical boundary in this case. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
Because a city licensing office receives abbreviations that map to different local agencies depending on context, “Successful provisioning by itself proves all security, output, and operational constraints are satisfied” does not fully explain the decisive technical boundary in this case. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
“The most powerful or least restricted model is always the safest choice regardless of task conditions” does not fully explain the decisive technical boundary in this case. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
Question 778
Production requirement: A city licensing office receives abbreviations that map to different local agencies depending on context. Which evidence is strongest for proving this requirement was met?
Explanation
Correct Answer
A city licensing office receives abbreviations that map to different local agencies depending on context. A city licensing office receives abbreviations that map to different local agencies depending on context. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Incorrect Answers
Because a city licensing office receives abbreviations that map to different local agencies depending on context, “The team closes a change ticket without replaying a representative positive and negative test” alone does not establish all the required behavior. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
“A project dashboard remains green, although the affected request, output and permissions were not checked” would not meet the requirement. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
“The operator reports success without reproducing the observed behavior: a city licensing office receives abbreviations that map to different local agencies depending on context” alone does not establish all the required behavior. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
Question 779
A team is investigating this task: A city licensing office receives abbreviations that map to different local agencies depending on context. What targeted records should the responder inspect to distinguish configuration from runtime behavior?
Explanation
Correct Answer
Because a city licensing office receives abbreviations that map to different local agencies depending on context, compare source offsets, label definitions, confidence or evidence, schema errors and human adjudication examples. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Incorrect Answers
A city licensing office receives abbreviations that map to different local agencies depending on context. “Restart every affected component before collecting request or tool traces” would not be the most direct first check for this failure. Compare source offsets, label definitions, confidence or evidence, schema errors and human adjudication examples.
A city licensing office receives abbreviations that map to different local agencies depending on context. “Read only the broad service-health summary and skip actual workload evidence” would not meet the requirement. Compare source offsets, label definitions, confidence or evidence, schema errors and human adjudication examples. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
“Assume the general service is at fault without checking the affected operation in this case: a city licensing office receives abbreviations that map to different local agencies depending on context” would not meet the requirement. Compare source offsets, label definitions, confidence or evidence, schema errors and human adjudication examples.
Question 780
Situation: A city licensing office receives abbreviations that map to different local agencies depending on context. Which TWO proposed practices create material safety, correctness, or reliability risks? Select TWO answers.
Explanation
Correct Answer
Because a city licensing office receives abbreviations that map to different local agencies depending on context, “Treat every capitalized word as an organization entity” would violate the requirement. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
“Invent missing account numbers to satisfy a required field” would violate the requirement. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
Incorrect Answers
“Inspect records that explain the affected process, then check the result described in this case: A city licensing office receives abbreviations that map to different local agencies depending on context” adds a relevant safeguard or verification step rather than weakening the solution. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain.
“Define the allowed entity schema and normalization rules, validate extracted spans against source text and route ambiguous cases for review” is the controlled implementation, not an unsafe workaround. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Confirm that extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Because a city licensing office receives abbreviations that map to different local agencies depending on context, “Extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented” is the required outcome of a working control, not an action that weakens it. Entity labels are not reliable unless type definitions, span evidence and normalization behavior are consistent across the domain. Confirm that extracted entities match source spans, required categories use approved canonical values and ambiguous mentions are flagged rather than invented.
Developing AI Apps and Agents on Azure (AI-103) Practice Tests
- Microsoft AI-103 Practice Test 76 - 10 Questions (Last Reviewed 03-Nov-2026)
- Microsoft AI-103 Practice Test 77 - 10 Questions (Last Reviewed 02-Nov-2026)
- Microsoft AI-103 Practice Test 78 - 10 Questions (Last Reviewed 01-Nov-2026)
- Microsoft AI-103 Practice Test 79 - 10 Questions (Last Reviewed 31-Oct-2026)
- Microsoft AI-103 Practice Test 80 - 10 Questions (Last Reviewed 30-Oct-2026)