Microsoft Security Exposure Management
Vulnerabilities and exposurePredefined classifications in Microsoft Security Exposure Management
In brief
The page title and description now identify it as a list of current predefined classifications. Its heading was also simplified.
What Defender admins need to know
Administrators can use the updated page labeling to locate classification information more easily.
Summaries are generated from the documentation change itself.
Documentation change
View on Microsoft LearnThe comparison below shows only the changed extract. Use the full-page view for complete context.
Predefined classifications - Microsoft Security Research
Microsoft Security Exposure Management keeps your business-critical assets secure and available. Critical assets help the SOC team prioritize efforts to maximize the impact on improving the organization's security posture. This article outlines the built-in critical asset classifiers provided by default, which are automatically updated by Microsoft Security Research teams.
| Classification | Asset type | Default criticality level | Description |
|---|---|---|---|
| Executive-Sponsored AI Agent | AI agent | Medium | This rule applies to AI agents that are created or owned by senior executives in the organization. As such, these agents may be granted access to sensitive data and act on the executive's behalf. Compromise could lead to unauthorized actions performed under executive authority, and exposure of sensitive executive-level data - without the executive's identity being directly compromised. |
@@ -1,6 +1,6 @@ ----title: Criticality Levels for Classifications-description: Lists of all the current criticality levels for classifications in Microsoft Security Exposure Management.+title: Predefined classifications in Microsoft Security Exposure Management+description: Lists of all the current predefined classifications in Microsoft Security Exposure Management. ms.topic: reference ms.author: dlanger author: dlanger@@ -10,7 +10,7 @@ ms.custom: sfi-ga-nochange ai-usage: ai-assisted --- -# Predefined classifications - Microsoft Security Research+# Predefined classifications Microsoft Security Exposure Management keeps your business-critical assets secure and available. Critical assets help the SOC team prioritize efforts to maximize the impact on improving the organization's security posture. This article outlines the built-in critical asset classifiers provided by default, which are automatically updated by Microsoft Security Research teams. @@ -168,6 +168,6 @@ Current asset types are: | Classification | Asset type | Default criticality level | Description | | -------------- | ---------- | ------------------------- | ----------- | | Executive-Sponsored AI Agent | AI agent | Medium | This rule applies to AI agents that are created or owned by senior executives in the organization. As such, these agents may be granted access to sensitive data and act on the executive's behalf. Compromise could lead to unauthorized actions performed under executive authority, and exposure of sensitive executive-level data - without the executive's identity being directly compromised. |-| AI Agent with Privileged Business System Write Access | AI agent | Medium | This rule applies to AI agents configured with tools that can perform high-risk write operations on business-critical systems. These operations include creating, modifying, and deleting records such as sales orders, customer data, financial transactions, and legal agreements. Compromise could lead to significant business impact. |+