AIF-C01 Applications of Foundation Models Practice Question

A company uses Amazon Bedrock to generate product descriptions. They want to ensure the outputs consistently follow a specific brand tone (professional yet friendly). They have a small set of example descriptions (few-shot examples) but do not want to fine-tune the model. Which strategy best achieves consistent tone without modifying the base model?

⚠ Common exam trap

The trap is thinking that fine-tuning is required for consistent tone, but few-shot prompting with a system prompt is often sufficient and avoids model modification.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

Use a system prompt that defines the brand tone and include few-shot examples in the prompt.

Using a system prompt to define the brand tone and including few-shot examples in the prompt is the best strategy because it guides the model's behavior without modifying its weights. This leverages in-context learning, which is effective for style adaptation. Fine-tuning is unnecessary and more resource-intensive.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
Fine-tune the model on a dataset of product descriptions that exemplify the desired tone.
Why it's wrong here
Fine-tuning modifies the base model's weights, which the stem explicitly excludes; it also requires a far larger labelled dataset than the few-shot examples available. It is tempting because fine-tuning strongly shapes style, and would be correct if the constraint against modifying the model were absent and ample training data existed.
✓
Use a system prompt that defines the brand tone and include few-shot examples in the prompt.
Why this is correct
A system prompt sets persistent behavioural instructions, while few-shot examples demonstrate the desired professional-yet-friendly phrasing in context. Together they steer generation at inference time, achieving consistent brand tone without altering weights or incurring fine-tuning cost.
✗
Implement prompt chaining by breaking the task into multiple steps, each with its own prompt.
Why it's wrong here
Prompt chaining sequences separate prompts for multi-step tasks, but tone adherence is a single-generation constraint, not a decomposition problem. Splitting the request adds latency and drift without enforcing style. It suits workflows needing intermediate reasoning or tool calls between stages, whereas few-shot examples embedded in one prompt directly anchor the brand voice.
✗
Use retrieval-augmented generation (RAG) to pull example descriptions from a database and prepend them to the prompt.
Why it's wrong here
RAG retrieves passages by semantic similarity to the input, so the examples injected vary per request and cannot guarantee a fixed tone. It suits grounding answers in a large, changing knowledge corpus. Few-shot prompting places the same curated examples in every prompt, giving the consistent brand voice required.

Go deeper

AIF-C01 Applications of Foundation Models Practice Question

Why each option matters

Option-by-option breakdown

Related to this question