A company uses an Amazon Bedrock model for a document-summarization API. Each request sends a 10,000-token document and receives a 500-token summary. The team wants to reduce inference cost per request. Which change is most effective?

Chunk the document and summarize each section, then combine the summaries

Switch to a larger model for better summarization quality

Increase the max output token limit to 2,000

Send the document twice to ensure the model captures all details

A healthcare startup wants to build a model that generates clinical notes from doctor-patient conversations. They have a small dataset of 5,000 de-identified transcripts. The team needs the model to understand medical terminology and produce structured notes. Which lifecycle approach is most appropriate?

Pre-train a new transformer model from scratch on the 5,000 transcripts

Fine-tune the model on a general text corpus before using it

Use the foundation model as-is without any additional training

Start with a pre-trained foundation model and fine-tune it on the clinical transcripts

In the context of large language models, what is a token?

A prompt template that instructs the model to generate a specific type of response

A fixed-size vector that represents the semantic meaning of a word or phrase

A chunk of text that is split into smaller pieces to fit within the model's context window

A unit of text, such as a word, part of a word, or punctuation mark, that the model processes as a single element

A developer is building a search tool that finds similar legal cases based on a query. The tool must return cases that are semantically similar, not just keyword matches. Which approach should the developer use?

Use a speech-to-text model to transcribe the query

Use a diffusion model to generate similar case summaries

Use a keyword search that matches exact terms in the query

Convert the query and cases into embeddings and compute vector similarity

AWSCertified AI Practitioner

Domain 2Objective 1

Task Statement 2.1: Explain the Basic Concepts of Generative AI (GenAI). AIF-C01 Practice Questions (Page 5)

Part of the Content Domain 2: Fundamentals of GenAI domain, which makes up ~19% of our current practice bank. AWS does not publish an official question count, but from its 90-minute exam (~35–60 total, ~7–11 in this domain), expect 2–4 from this objective — we provide 32 practice questions to prepare you well beyond it. (estimate)

32questions here

7free pages

6concepts

Questions 21–25

21application · medium
A company uses an Amazon Bedrock model for a document-summarization API. Each request sends a 10,000-token document and receives a 500-token summary. The team wants to reduce inference cost per request. Which change is most effective?
Select an answer first
22application · medium
A healthcare startup wants to build a model that generates clinical notes from doctor-patient conversations. They have a small dataset of 5,000 de-identified transcripts. The team needs the model to understand medical terminology and produce structured notes. Which lifecycle approach is most appropriate?
Select an answer first
23foundation · easy
During which stage of the foundation model lifecycle would you collect and curate a large, diverse dataset to train the model on general language understanding?
Select an answer first
24foundation · easy
In the context of large language models, what is a token?
Select an answer first
25application · medium
A developer is building a search tool that finds similar legal cases based on a query. The tool must return cases that are semantically similar, not just keyword matches. Which approach should the developer use?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to AIF-C01

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by AWS. “AIF-C01” is a trademark of its owner, used for identification only.

Task Statement 2.1: Explain the Basic Concepts of Generative AI (GenAI). AIF-C01 Practice Questions (Page 5)

Questions 21–25

A company uses an Amazon Bedrock model for a document-summarization API. Each request sends a 10,000-token document and receives a 500-token summary. The team wants to reduce inference cost per request. Which change is most effective?

During which stage of the foundation model lifecycle would you collect and curate a large, diverse dataset to train the model on general language understanding?

In the context of large language models, what is a token?

A developer is building a search tool that finds similar legal cases based on a query. The tool must return cases that are semantically similar, not just keyword matches. Which approach should the developer use?

Keep studying — Content Domain 2: Fundamentals of GenAI