AWS amazonq medium security documentation change

Service: amazonq · 2025-04-11 · Security-related medium

File: amazonq/latest/qbusiness-ug/cw-log-examples.md

Summary

Added examples of Amazon Q Business chat logs with hallucination detection fields and mitigation status descriptions

Security assessment

Introduces logging fields and documentation for hallucination mitigation, which addresses AI response accuracy and misinformation risks

Diff

diff --git a/amazonq/latest/qbusiness-ug/cw-log-examples.md b/amazonq/latest/qbusiness-ug/cw-log-examples.md
index 9b25d672f..214564aeb 100644
--- a//amazonq/latest/qbusiness-ug/cw-log-examples.md
+++ b//amazonq/latest/qbusiness-ug/cw-log-examples.md
@@ -26,0 +27,51 @@ The following are examples of Amazon Q Business chat message and feedback logs i
+**Amazon Q Business generated chat message (no hallucination detected)**
+    
+    
+    {
+        "hallucinated_message": "string",
+        "application_id": "<ApplicationId>",
+        "event_timestamp": "2024-09-05T19:43:58Z",
+        "log_type": "Message",
+        "account_id": "<AccountId>",
+        "conversation_id": "<conversationId>",
+        "system_message_id": "<systemMessageId>",
+        "user_message_id": "<userMessageId>",
+        "user_message": string,
+        "system_message": string,
+        "output_metrics_is_message_blocked": boolean,
+        "output_metrics_is_message_with_no_answer": boolean,
+        "user_email": "<userEmail>"
+    }
+
+**Amazon Q Business generated chat message (hallucination detected)**
+    
+    
+    {
+        "hallucinated_message": "NO HALLUCINATION DETECTED",
+        "application_id": "<ApplicationId>",
+        "event_timestamp": "2024-09-05T19:43:58Z",
+        "log_type": "Message",
+        "account_id": "<AccountId>",
+        "conversation_id": "<conversationId>",
+        "system_message_id": "<systemMessageId>",
+        "user_message_id": "<userMessageId>",
+        "user_message": string,
+        "system_message": string,
+        "output_metrics_is_message_blocked": boolean,
+        "output_metrics_is_message_with_no_answer": boolean,
+        "user_email": "<userEmail>"
+    }
+
+For system generated messages that have hallucination, you'll see one of the following log descriptions:
+
+  * Hallucination mitigation disabled – `hallucinated_message: DISABLED`
+
+  * Hallucination mitigation enabled but not triggered – `hallucinated_message: NOT TRIGGERED`
+
+  * Hallucination mitigation enabled and triggered, but no hallucinations detected – `hallucinated_message: NO HALLUCINATION DETECTED`
+
+  * Hallucination mitigation feature enabled and triggered, and hallucinations detected – `hallucinated_message: `string``
+
+
+
+