AIF-C01 Fundamentals of Generative AI Practice Question
A developer is using the Amazon Bedrock API to generate text. They notice that the model sometimes returns harmful content despite setting safety parameters. What is the BEST way to add an additional layer of content filtering?
⚠ Common exam trap
The AIF-C01 exam often tests the misconception that fine-tuning or prompt engineering alone can fully prevent harmful outputs, when in fact a separate, configurable guardrail layer is the recommended approach for production-grade content filtering in Amazon Bedrock.
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
✓
Configure content filters in Amazon Bedrock Guardrails
Amazon Bedrock Guardrails provides a dedicated, configurable content filtering layer that can block harmful content at inference time, independent of the model's built-in safety parameters. This allows developers to enforce custom policies (e.g., hate speech, violence) without modifying the model itself, making it the best additional safeguard.
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 curated safe dataset
Why it's wrong here
Fine-tuning alters model weights for domain behaviour, not runtime content enforcement, and cannot guarantee harmful completions are blocked. It is tempting because curated safe data reduces harmful tendencies during training, and would be correct when adapting a model to a specialised domain or tone rather than adding a filtering layer.
- ✓
Configure content filters in Amazon Bedrock Guardrails
Why this is correct
Amazon Bedrock Guardrails applies configurable content filters that evaluate both prompts and model responses independently of the model's own safety parameters, blocking harmful categories the model may still emit and adding a deterministic policy layer.
- ✗
Improve prompt engineering with more specific instructions
Why it's wrong here
Prompt engineering shapes model behaviour but provides no deterministic post-generation filter; a sufficiently adversarial prompt can still elicit harmful output. It is tempting because refining instructions is the first-line technique for steering Bedrock responses, and would be correct when the goal is improving output relevance rather than enforcing a hard content boundary.
- ✗
Use AWS WAF to filter API responses
Why it's wrong here
AWS WAF inspects inbound HTTP requests at the edge; it cannot parse or filter the natural-language content returned by Bedrock, so harmful model output passes through untouched. It is tempting because WAF is the standard AWS control for blocking malicious web traffic, and would be correct for filtering inbound request payloads.
Go deeper
Related to this question
About these practice questions
One of 862 original AIF-C01 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AIF-C01 practice question is part of Courseiva's free Amazon Web Services certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AIF-C01 exam.