Curriculum Overview: Securing Generative AI Applications on AWS

This curriculum focuses on the critical security task of implementing protections and guardrails for Generative AI (GenAI) applications within the AWS ecosystem, specifically aligned with the AWS Certified Security - Specialty (SCS-C03) exam requirements (Skill 3.2.7).

Prerequisites

Before starting this module, learners should have a firm grasp of the following:

AWS Identity and Access Management (IAM): Proficiency in creating least-privilege policies and service roles for compute workloads.
Infrastructure Security: Knowledge of AWS WAF, Security Groups, and Network ACLs.
Cloud Fundamentals: Understanding of Amazon EC2, Lambda, and basic container orchestration.
AI/ML Basics: Conceptual understanding of Large Language Models (LLMs) and the typical request/response flow in a GenAI application.

Module Breakdown

This curriculum is divided into four focused modules, progressing from threat identification to technical implementation.

Module	Topic	Complexity	Focus Area
1	The GenAI Threat Landscape	Intermediate	OWASP Top 10 for LLM Applications
2	Input Protections	Advanced	Prompt Injection & Input Filtering
3	Model & Output Guardrails	Advanced	Bedrock Guardrails & Content Filtering
4	Monitoring & Governance	Intermediate	Logging, Auditing, and Compliance

Learning Objectives per Module

Module 1: The GenAI Threat Landscape

Identify and categorize the OWASP Top 10 for LLM Applications (e.g., LLM01: Prompt Injection, LLM02: Insecure Output Handling).
Understand the