Merge pull request #22810 from github/jacknojo-llm-output-threat-model

Add opt-in AI output threat model

Author: jacknojo

docs/codeql/reusables/threat-model-description.rst

@@ -11,6 +11,7 @@ Note that subcategories can be turned included or excluded separately, so you ca
 
 The less commonly used categories are:
 
+- ``ai`` which represents output generated by large language models (LLMs). This threat model is opt-in.
 - ``android`` which represents reads from external files in Android (``android-external-storage-dir``) and parameter of an entry-point method declared in a ``ContentProvider`` class (``contentprovider``). Currently only used by Java/Kotlin.
 - ``database-access-result`` which represents a database access. Currently only used by JavaScript.
 - ``file-write`` which represents opening a file in write mode. Currently only used in C#.

javascript/ql/test/library-tests/threat-models/ai-enabled/ActiveAiThreatModel.expected

@@ -0,0 +1 @@
+| ai |

javascript/ql/test/library-tests/threat-models/ai-enabled/ActiveAiThreatModel.ext.yml

@@ -0,0 +1,6 @@
+extensions:
+  - addsTo:
+      pack: codeql/threat-models
+      extensible: threatModelConfiguration
+    data:
+      - ["ai", true, 0]

javascript/ql/test/library-tests/threat-models/ai-enabled/ActiveAiThreatModel.ql

@@ -0,0 +1,6 @@
+private import codeql.threatmodels.ThreatModels
+
+where
+  knownThreatModel("ai") and
+  currentThreatModel("ai")
+select "ai"

shared/threat-models/change-notes/2026-10-12-ai-threat-model.md

@@ -0,0 +1,4 @@
+---
+category: feature
+---
+* Added the opt-in `ai` threat model for LLM-generated data. No built-in source models are included.

shared/threat-models/ext/threat-model-grouping.model.yml

@@ -30,3 +30,5 @@ extensions:
       - ["file-write", "all"]
       - ["reverse-dns", "all"]
       - ["view-component-input", "all"]
+      # AI threat models
+      - ["ai", "all"]