AIF-C01 Applications of Foundation Models Practice Question

A financial services company uses Amazon Bedrock with the Anthropic Claude 3 Sonnet model to answer employee questions about internal policies. The policy documents are updated frequently, and the model occasionally provides outdated or incorrect policy details. The company wants the model to base its answers on the most current authoritative documents without retraining the model. Which approach should they use?

⚠ Common exam trap

The trap here is assuming that a larger or fine-tuned model automatically knows frequently updated internal documents, when grounding requires retrieval of current source content at inference time.

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 Retrieval Augmented Generation (RAG) by storing policy documents in an Amazon Bedrock knowledge base and retrieving relevant passages at inference time.

Retrieval Augmented Generation with an Amazon Bedrock knowledge base retrieves relevant, current policy passages and supplies them to the model as context, so answers are grounded in authoritative documents. It avoids retraining costs and keeps pace with frequent updates, directly solving the outdated or incorrect policy detail problem.

Answer analysis

Option-by-option breakdown

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

✗
Switch to a larger foundation model with a higher parameter count.
Why it's wrong here
A larger model may have stronger general reasoning, but it still lacks access to the company's frequently updated internal policy documents. Model size does not solve staleness or grounding; without retrieval or fine-tuning, the model relies on its pretraining knowledge. The scenario requires current authoritative content, which a larger model alone cannot provide.
✓
Use Retrieval Augmented Generation (RAG) by storing policy documents in an Amazon Bedrock knowledge base and retrieving relevant passages at inference time.
Why this is correct
RAG with an Amazon Bedrock knowledge base indexes the policy documents and retrieves the most relevant passages for each query, then passes them to the model as context. This grounds responses in the latest authoritative content without retraining, and updating the knowledge base data source keeps answers current. It directly addresses outdated or incorrect policy details by supplying fresh source text at inference time.
✗
Fine-tune the foundation model on the updated policy documents.
Why it's wrong here
Fine-tuning changes the model's weights and requires a labeled dataset and a training job, which is expensive and slow for documents that change frequently. It also risks catastrophic forgetting and does not provide per-query grounding. For a policy Q&A system needing fresh, authoritative answers, retrieval-augmented generation is a better fit because it injects the current documents at inference time.
✗
Increase the model's temperature setting to encourage more creative and up-to-date answers.
Why it's wrong here
Temperature controls randomness in token selection; increasing it makes outputs more varied, not more accurate or current. Higher temperature can increase hallucinations, which is the opposite of what a compliance-focused policy assistant needs. This setting does not give the model access to updated documents and cannot ensure answers reflect the latest authoritative policy text.

Go deeper

AIF-C01 Applications of Foundation Models Practice Question

Why each option matters

Option-by-option breakdown

Related to this question