AIShield - Image classification AI Model Evasion low suspicious vulnerability detection

Back


Id	c4bedb3c-4fb8-4b1c-af5b-8229bd25f521
Rulename	AIShield - Image classification AI Model Evasion low suspicious vulnerability detection
Description	This alert creates an incident when Image classification AI Model Evasion low suspicious, high severity vulnerability detected from the AIShield.
Severity	High
Required data connectors	BoschAIShield
Kind	Scheduled
Query frequency	1h
Query period	1h
Trigger threshold	0
Trigger operator	gt
Source Uri	https://github.com/Azure/Azure-Sentinel/blob/master/Solutions/AIShield AI Security Monitoring/Analytic Rules/ImageClassficationModelEvasionLowSuspiciousVulnDetection.yaml
Version	1.0.4
Arm template	c4bedb3c-4fb8-4b1c-af5b-8229bd25f521.json

KQL

AIShield
| where Message has 'Image Classification AI Model Evasion Attack Identified'
| where Severity =~ 'Medium'
| where SuspiciousLevel =~ 'low suspicious attack'
| extend NTDomain = tostring(split(Computer, '\\')[0]), HostName = tostring(split(Computer, '.')[0])

YAML

triggerOperator: gt
queryFrequency: 1h
requiredDataConnectors:
- connectorId: BoschAIShield
  dataTypes:
  - AIShield
relevantTechniques: []
entityMappings:
- entityType: Host
  fieldMappings:
  - identifier: HostName
    columnName: HostName
  - identifier: NTDomain
    columnName: NTDomain
- entityType: IP
  fieldMappings:
  - identifier: Address
    columnName: SourceName
query: |
  AIShield
  | where Message has 'Image Classification AI Model Evasion Attack Identified'
  | where Severity =~ 'Medium'
  | where SuspiciousLevel =~ 'low suspicious attack'
  | extend NTDomain = tostring(split(Computer, '\\')[0]), HostName = tostring(split(Computer, '.')[0])  
triggerThreshold: 0
alertDetailsOverride:
  alertDisplayNameFormat: AIShield - Image Classification AI Model Evasion low suspicious vulnerability detected.
  alertSeverityColumnName: Severity
  alertTacticsColumnName: 
  alertDescriptionFormat: |
        This query detects Image Classification AI Model Evasion low suspicious, medium severity alert from AIShield generated at {{TimeGenerated}}.\n\nPlease check the source for more information and investigate further.
OriginalUri: https://github.com/Azure/Azure-Sentinel/blob/master/Solutions/AIShield AI Security Monitoring/Analytic Rules/ImageClassficationModelEvasionLowSuspiciousVulnDetection.yaml
queryPeriod: 1h
name: AIShield - Image classification AI Model Evasion low suspicious vulnerability detection
status: Available
kind: Scheduled
description: |
    'This alert creates an incident when Image classification AI Model Evasion low suspicious, high severity vulnerability detected from the AIShield.'
id: c4bedb3c-4fb8-4b1c-af5b-8229bd25f521
version: 1.0.4
eventGroupingSettings:
  aggregationKind: SingleAlert
tactics: []
severity: High

ARM