AWS aws-certification high security documentation change

Service: aws-certification · 2026-05-01 · Security-related high

File: aws-certification/latest/ai-practitioner-01/ai-practitioner-01-domain5.md

Summary

Added new security features (Bedrock AgentCore Identity, Policy, Guardrails), expanded security/privacy considerations (data leakage prevention, output validation, audit logging), and added hallucination detection methods.

Security assessment

Directly adds documentation for AI security features (Guardrails, data leakage prevention) and addresses security vulnerabilities like prompt injection and hallucination risks through mitigation techniques (output validation, RAG grounding).

Diff

diff --git a/aws-certification/latest/ai-practitioner-01/ai-practitioner-01-domain5.md b/aws-certification/latest/ai-practitioner-01/ai-practitioner-01-domain5.md
index 3d426d9b0..8090e4916 100644
--- a//aws-certification/latest/ai-practitioner-01/ai-practitioner-01-domain5.md
+++ b//aws-certification/latest/ai-practitioner-01/ai-practitioner-01-domain5.md
@@ -26 +26 @@ Objectives:
-  * Identify AWS services and features to secure AI systems (for example, IAM roles, policies, and permissions; encryption; Amazon Macie; AWS PrivateLink; AWS shared responsibility model).
+  * Identify AWS services and features to secure AI systems (for example, IAM roles, policies, and permissions; encryption; Amazon Macie; AWS PrivateLink; AWS shared responsibility model; Amazon Bedrock AgentCore Identity; Policy in AgentCore; Amazon Bedrock Guardrails).
@@ -32 +32,3 @@ Objectives:
-  * Describe security and privacy considerations for AI systems (for example, application security, threat detection, vulnerability management, infrastructure protection, prompt injection, encryption at rest and in transit).
+  * Describe security and privacy considerations for AI systems (for example, application security, threat detection, vulnerability management, infrastructure protection, prompt injection, encryption at rest and in transit, data leakage prevention, output filtering and validation, audit trail and logging requirements for AI interactions, toxicity).
+
+  * Describe hallucination detection methods and grounding techniques to improve output accuracy (for example, Retrieval Augmented Generation [RAG] grounding, output validation, confidence scoring).
@@ -58 +60 @@ Content Domain 4: Guidelines for Responsible AI
-In-scope AWS services and features
+In-Scope AWS Services