AWS wellarchitected medium security documentation change

Service: wellarchitected · 2026-01-28 · Security-related medium

File: wellarchitected/latest/financial-services-industry-lens/fsisec07.md

Summary

Extended network monitoring guidance to include AI endpoint access, model invocations, and data flow patterns for generative AI services.

Security assessment

The change adds security-focused monitoring for AI endpoints and data flows, evidenced by 'monitoring AI endpoint access and authentication attempts' and 'tracking data access across AI systems', addressing potential network security risks in AI deployments.

Diff

diff --git a/wellarchitected/latest/financial-services-industry-lens/fsisec07.md b/wellarchitected/latest/financial-services-industry-lens/fsisec07.md
index 5de6c7966..eebf1d3da 100644
--- a//wellarchitected/latest/financial-services-industry-lens/fsisec07.md
+++ b//wellarchitected/latest/financial-services-industry-lens/fsisec07.md
@@ -9 +9 @@ FSISEC07-BP01 Monitor instance trafficFSISEC07-BP02 Use VPC Traffic MirroringFSI
-Monitor network traffic for expected and unexpected traffic to identify irregularities and gain key insights into the security of the system. For example, a poorly-performing network can indicate that the network is under threat, and irregular attempts to contact unexpected external systems can indicate that an internal host has been compromised. 
+Monitor network traffic for expected and unexpected traffic to identify irregularities and gain key insights into the security of the system. For example, a poorly-performing network can indicate that the network is under threat, and irregular attempts to contact unexpected external systems can indicate that an internal host has been compromised. With generative AI services, inspection includes monitoring AI endpoint access and authentication attempts, model invocations, and data flow patterns. 
@@ -14,0 +15,2 @@ Amazon EC2 instances automatically track aggregate network inbound and outbound
+For AI workloads, implement comprehensive monitoring of model endpoint access and API usage patterns while establishing private network communication and tracking data access across AI systems. 
+