Unusual AI model invocation

Cortex Toolbox Ingest Calculator Platform Changes

Detectors ▾

All detectors Statistics ATT&CK matrix

Analytics Alerts BIOCs Correlation Rules

Content ▾

Packs Integrations MCP Servers Threat Feeds Reputation Commands

Automation

Playbooks Scripts Jobs Automation Rules Lists

Case and Issue Setup

Incident Fields Layouts Layout Rules

Threat Intelligence

Indicator Types Indicator Fields Reputation Commands Threat Feeds

Data Ingestion

Parsing Rules Modeling Rules XDRC Templates

Dashboards & Reports

XSIAM Dashboards XSIAM Reports

Docs ▾

Documentation XDM Explorer Cortex Versions

Settings ▾

Sign in

Analytics BIOC Informational

✕

Unusual AI model invocation

A cloud identity invoked an AI model for the first time. MITRE ATLAS Technique: AML.T0050 - Command and Scripting Interpreter.

Module:: Cortex Cloud
Licensed by:: Cloud Runtime Security (CRS)
Data source:: AWS Audit Log, Gcp Audit Log

ATT&CK tactics: Execution (TA0002)

ATT&CK techniques: Command and Scripting Interpreter: Cloud API (T1059.009)

Detector tags: Cloud AI Infrastructure Analytics

Attacker's goals:

Gain access to AI models.

Investigative actions:

Examine which AI models were invoked. Investigate any unusual activity originating from the suspected identity.

Test period:: N/A (single event)
Deduplication:: 5 Days