AWS Certified AI Practitioner AIF-C01 Practice Test 22
Question 210
Several specialized agents must research claims, calculate a payment and request final human approval. After selecting the recommended approach, which evidence MOST DIRECTLY demonstrates that the actual required outcome has been achieved?
- A system inventory lists this competing approach as active: Grant every agent unrestricted payout permissions
- A component test invokes this alternate approach successfully: Use isolated chatbots with no handoff state
- A reviewed design record approves this competing approach: Use embeddings search as a final claim settlement
- Audit traces show subtask results and an approval gate before payout
Explanation
Correct Answer
D:Audit traces show subtask results and an approval gate before payout. This checks the business result rather than just a configured feature.
Incorrect Answers
A:Unrestricted write access bypasses human approval An active setting cannot independently prove the requested business outcome.
B:Disconnected assistants cannot reliably compose the process A component invocation does not validate the necessary end-to-end outcome.
C:Retrieval alone does not authorize or execute payment Design approval is not evidence that the alternative meets the actual requirement.
Question 211
A team is calculating whether AI summaries save analyst time. After selecting the recommended approach, which evidence MOST DIRECTLY demonstrates that the actual required outcome has been achieved?
- A component test invokes this alternate approach successfully: Use prompt length as productivity gained
- A controlled test reports accepted summaries and verified reviewer hours saved
- A reviewed design record approves this competing approach: Assume context capacity alone means positive ROI
- A system inventory lists this competing approach as active: Count summaries produced without reviewing quality
Explanation
Correct Answer
B:A controlled test reports accepted summaries and verified reviewer hours saved. This checks the business result rather than just a configured feature.
Incorrect Answers
A:Token volume is not staff time saved A component invocation does not validate the necessary end-to-end outcome.
C:Capacity does not measure business value Design approval is not evidence that the alternative meets the actual requirement.
D:Volume can rise while errors create more work An active setting cannot independently prove the requested business outcome.
Question 212
An internal search tool must find semantically related policies even when the wording differs. Which option BEST meets the business requirement?
- Rely only on exact filename search
- Compare character counts as semantic relevance
- Use sampling temperature as an index
- Generate embeddings and use vector similarity
Explanation
Correct Answer
D:Embedding vectors represent useful semantic relationships The specified scenario makes this the most appropriate approach.
Incorrect Answers
A:Names miss paraphrased content This prevents the approach from meeting the stated requirement.
B:Documents of equal length may be unrelated This prevents the approach from meeting the stated requirement.
C:Temperature controls generation rather than retrieval This prevents the approach from meeting the stated requirement.
Question 213
An agent must access approved external APIs through a common tool interface. A pilot uses this approach instead: Use tokenization as the API authorization mechanism. Which limitation MOST DIRECTLY explains why the requirement is not met?
- Prompt text cannot establish secure tool registration
- Breaking text into tokens does not authenticate API calls
- Similarity is not authorization
- MCP standardizes tool connectivity but does not grant permission automatically
Explanation
Correct Answer
B:This is the actual failure mode of the proposed approach: Use tokenization as the API authorization mechanism. Breaking text into tokens does not authenticate API calls
Incorrect Answers
A:This limitation concerns Assume mentioning a tool in a prompt makes it callable; it does not diagnose the proposed Use tokenization as the API authorization mechanism approach.
C:This limitation concerns Use vector relevance as permission to transfer money; it does not diagnose the proposed Use tokenization as the API authorization mechanism approach.
D:MCP standardizes tool connectivity but does not grant permission automatically This explains the recommended design, but it is not the requested diagnosis of Use tokenization as the API authorization mechanism.
Question 214
Several specialized agents must research claims, calculate a payment and request final human approval. Before approving a rollout, an architect suggests Use embeddings search as a final claim settlement. Which mismatch should be documented for this precise design?
- Coordinated agents can combine outputs while preserving authority checks
- Unrestricted write access bypasses human approval
- Disconnected assistants cannot reliably compose the process
- Retrieval alone does not authorize or execute payment
Explanation
Correct Answer
D:This is the actual failure mode of the proposed approach: Use embeddings search as a final claim settlement. Retrieval alone does not authorize or execute payment
Incorrect Answers
A:Coordinated agents can combine outputs while preserving authority checks This explains the recommended design, but it is not the requested diagnosis of Use embeddings search as a final claim settlement.
B:This limitation concerns Grant every agent unrestricted payout permissions; it does not diagnose the proposed Use embeddings search as a final claim settlement approach.
C:This limitation concerns Use isolated chatbots with no handoff state; it does not diagnose the proposed Use embeddings search as a final claim settlement approach.
Question 215
A team is calculating whether AI summaries save analyst time. A pilot uses this approach instead: Count summaries produced without reviewing quality. Which limitation MOST DIRECTLY explains why the requirement is not met?
- Token volume is not staff time saved
- Time saved must be balanced against factual correction work
- Volume can rise while errors create more work
- Capacity does not measure business value
Explanation
Correct Answer
C:This is the actual failure mode of the proposed approach: Count summaries produced without reviewing quality. Volume can rise while errors create more work
Incorrect Answers
A:This limitation concerns Use prompt length as productivity gained; it does not diagnose the proposed Count summaries produced without reviewing quality approach.
B:Time saved must be balanced against factual correction work This explains the recommended design, but it is not the requested diagnosis of Count summaries produced without reviewing quality.
D:This limitation concerns Assume context capacity alone means positive ROI; it does not diagnose the proposed Count summaries produced without reviewing quality approach.
Question 216
A retailer wants to show whether an assistant increases useful customer outcomes. A pilot uses this approach instead: Use tokens generated per second as the ROI metric. Which limitation MOST DIRECTLY explains why the requirement is not met?
- Model size is not a financial outcome
- Training accuracy does not capture customer conversations
- Speed alone does not measure customer resolution
- Business value requires outcome measures not only token throughput
Explanation
Correct Answer
C:This is the actual failure mode of the proposed approach: Use tokens generated per second as the ROI metric. Speed alone does not measure customer resolution
Incorrect Answers
A:This limitation concerns Treat parameter count as net revenue; it does not diagnose the proposed Use tokens generated per second as the ROI metric approach.
B:This limitation concerns Use model training accuracy as the sole business metric; it does not diagnose the proposed Use tokens generated per second as the ROI metric approach.
D:Business value requires outcome measures not only token throughput This explains the recommended design, but it is not the requested diagnosis of Use tokens generated per second as the ROI metric.
Question 217
An insurance assistant must interpret an uploaded photo together with a text instruction. Before approving a rollout, an architect suggests Use a filename-based regression score as the visual interpretation. Which mismatch should be documented for this precise design?
- Text-only input cannot directly inspect the photo
- Speech recognition does not process picture content
- File names are not evidence of damage in pixels
- Multimodal input processing covers both evidence types
Explanation
Correct Answer
C:This is the actual failure mode of the proposed approach: Use a filename-based regression score as the visual interpretation. File names are not evidence of damage in pixels
Incorrect Answers
A:This limitation concerns Use text-only inference without a visual description; it does not diagnose the proposed Use a filename-based regression score as the visual interpretation approach.
B:This limitation concerns Use audio transcription as image understanding; it does not diagnose the proposed Use a filename-based regression score as the visual interpretation approach.
D:Multimodal input processing covers both evidence types This explains the recommended design, but it is not the requested diagnosis of Use a filename-based regression score as the visual interpretation.
Question 218
A support agent must recall prior steps for the same customer without exposing other customers. A pilot uses this approach instead: Put every user's conversation in one unrestricted context. Which limitation MOST DIRECTLY explains why the requirement is not met?
- Shared raw history violates isolation
- No session state breaks follow-up references
- Sampling randomness cannot enforce access control
- Session-scoped memory maintains conversational continuity without cross-user leakage
Explanation
Correct Answer
A:This is the actual failure mode of the proposed approach: Put every user's conversation in one unrestricted context. Shared raw history violates isolation
Incorrect Answers
B:This limitation concerns Discard all prior context despite multi-turn requirements; it does not diagnose the proposed Put every user's conversation in one unrestricted context approach.
C:This limitation concerns Use model temperature to isolate sessions; it does not diagnose the proposed Put every user's conversation in one unrestricted context approach.
D:Session-scoped memory maintains conversational continuity without cross-user leakage This explains the recommended design, but it is not the requested diagnosis of Put every user's conversation in one unrestricted context.
Question 219
An engineer describes how a modern LLM relates distant words in a sentence. During a design review, someone proposes Treat speech transcription as the transformer architecture. Which technical objection is specifically relevant to this proposal?
- Rules do not capture learned contextual representations
- ASR is an application task, not the model architecture
- Transformer attention represents dependencies across token positions
- Retrieval indexes are external to in-model attention
Explanation
Correct Answer
B:This is the actual failure mode of the proposed approach: Treat speech transcription as the transformer architecture. ASR is an application task, not the model architecture
Incorrect Answers
A:This limitation concerns Describe an LLM as only a static rule table; it does not diagnose the proposed Treat speech transcription as the transformer architecture approach.
C:Transformer attention represents dependencies across token positions This explains the recommended design, but it is not the requested diagnosis of Treat speech transcription as the transformer architecture.
D:This limitation concerns Describe vector indexing as the whole network; it does not diagnose the proposed Treat speech transcription as the transformer architecture approach.
Question 220
A support agent must recall prior steps for the same customer without exposing other customers. Which option BEST meets the business requirement?
- Discard all prior context despite multi-turn requirements
- Put every user's conversation in one unrestricted context
- Use identity-scoped short-term session memory
- Use model temperature to isolate sessions
Explanation
Correct Answer
C:Session-scoped memory maintains conversational continuity without cross-user leakage The specified scenario makes this the most appropriate approach.
Incorrect Answers
A:No session state breaks follow-up references This prevents the approach from meeting the stated requirement.
B:Shared raw history violates isolation This prevents the approach from meeting the stated requirement.
D:Sampling randomness cannot enforce access control This prevents the approach from meeting the stated requirement.
