AIShield - AI Security Monitoring for Microsoft Sentinel

Solution: AIShield AI Security Monitoring

AIShield AI Security Monitoring Logo

Browse: 🏠 · Solutions · Connectors · Methods · Tables · Content · Parsers · ASIM Parsers · ASIM Products · 📊

↑ Back to Solutions Index

Attribute Value
Publisher AIShield
Support Tier Partner
Support Link https://azuremarketplace.microsoft.com/marketplace/apps/rbei.bgsw_aishield_product/
Categories domains
Version 3.0.1
Author AIShield - AIShield.Contact@bosch.com
First Published 2022-01-11
Last Updated 2025-03-06
Solution Folder AIShield AI Security Monitoring
Marketplace Azure Marketplace · Popularity: ⚪ Very Low (0%)

The AIShield AI Security Monitoring solution allows users to connect with AIShield custom defense mechanism logs with Microsoft Sentinel. It gives users more insight into their organization's AI assets security posturing and improves their AI systems security operation capabilities. AIShield.GuArdIan analyzes the LLM generated content to identify and mitigate harmful content, safeguarding against legal, policy, role based, and usage based violations.

Underlying Microsoft Technologies used:

This solution takes a dependency on the following technologies, and some of these dependencies either may be in Preview state or might result in additional ingestion or operational costs:

a. Azure Monitor HTTP Data Collector API

Contents

Data Connectors

This solution provides 1 data connector(s):

Tables Used

This solution uses 2 table(s):

Table Used By Connectors Used By Content
AIShield_CL AIShield Analytics, Workbooks
Guardian_CL 🔶 - Analytics

🔶 CLv1: This table uses the legacy Custom Log V1 schema format with type-suffixed column names (e.g. _s, _d, _b, _t, _g). Note: identification is based on column name suffixes which are also permitted in CLv2, so this classification may not always be accurate.

Content Items

This solution includes 42 content item(s):

Content Type Count
Analytic Rules 38
Workbooks 2
Parsers 2

Analytic Rules

Name Severity Tactics Tables Used
AIShield - Image Segmentation AI Model extraction high suspicious vulnerability detection High - AIShield_CL
AIShield - Image classification AI Model Evasion high suspicious vulnerability detection High - AIShield_CL
AIShield - Image classification AI Model Evasion low suspicious vulnerability detection High - AIShield_CL
AIShield - Image classification AI Model extraction high suspicious vulnerability detection High - AIShield_CL
AIShield - Natural language processing AI model extraction high suspicious vulnerability detection High - AIShield_CL
AIShield - Tabular classification AI Model Evasion high suspicious vulnerability detection High - AIShield_CL
AIShield - Tabular classification AI Model Evasion low suspicious vulnerability detection Medium - AIShield_CL
AIShield - Tabular classification AI Model extraction high suspicious vulnerability detection High - AIShield_CL
AIShield - Timeseries Forecasting AI Model extraction high suspicious vulnerability detection High - AIShield_CL
Guardian- Additional check JSON Policy Violation Detection Informational - Guardian_CL
Guardian- BII Detection Policy Violation Detection High - Guardian_CL
Guardian- Ban Topic Policy Violation Detection Medium - Guardian_CL
Guardian- Block Competitor Policy Violation Detection Low - Guardian_CL
Guardian- Blocks specific strings of text Policy Violation Detection Low - Guardian_CL
Guardian- Code Detection Policy Violation Detection Low - Guardian_CL
Guardian- Content Access Control Allowed List Policy Violation Detection Low - Guardian_CL
Guardian- Content Access Control Blocked List Policy Violation Detection Medium - Guardian_CL
Guardian- Content Safety Profanity Policy Violation Detection Low - Guardian_CL
Guardian- Content Safety Toxicity Policy Violation Detection. Low - Guardian_CL
Guardian- Gender Bias Policy Violation Detection Low - Guardian_CL
Guardian- Input Output Relevance Policy Violation Detection Informational - Guardian_CL
Guardian- Input Rate Limiter Policy Violation Detection Informational - Guardian_CL
Guardian- Invisible Text Policy Violation Detection Low - Guardian_CL
Guardian- Language Detection Policy Violation Detection Informational - Guardian_CL
Guardian- Malicious URL Policy Violation Detection Medium - Guardian_CL
Guardian- No LLM Output Policy Violation Detection Low - Guardian_CL
Guardian- Not Safe For Work Policy Violation Detection Low - Guardian_CL
Guardian- Privacy Protection PII Policy Violation Detection High - Guardian_CL
Guardian- Racial Bias Policy Violation Detection Low - Guardian_CL
Guardian- Regex Policy Violation Detection Low - Guardian_CL
Guardian- Same Input/Output Language Detection Policy Violation Detection Informational - Guardian_CL
Guardian- Secrets Policy Violation Detection Medium - Guardian_CL
Guardian- Security Integrity Checks Prompt Injection Policy Violation Detection High - Guardian_CL
Guardian- Sentiment Policy Violation Detection Low - Guardian_CL
Guardian- Special PII Detection Policy Violation Detection High - Guardian_CL
Guardian- Token Limit Policy Violation Detection Informational - Guardian_CL
Guardian- URL Detection Policy Violation Detection Informational - Guardian_CL
Guardian- URL Reachability Policy Violation Detection Informational - Guardian_CL

Workbooks

Name Tables Used
AIShield AIShield_CL
GuardianDashboard -

Parsers

Name Description Tables Used
AIShield - AIShield_CL (read)
Guardian - Guardian_CL (read)

Release Notes

Version Date Modified (DD-MM-YYYY) Change History
3.0.1 06-03-2025 Added new Analytic Rules.
Added new Workbook GuardianDashboard.json.
Added new Parser Guardian.yaml
3.0.0 15-01-2023 Added Entity Mapping and remove alertactics Column Name to Analytic Rules.
Added new Analytic Rules and updated existing Analytic Rules and (AIShield) Parser.

Browse: 🏠 · Solutions · Connectors · Methods · Tables · Content · Parsers · ASIM Parsers · ASIM Products · 📊

↑ Back to Solutions Index