0
1.0.3
Australia, Zurich
Standalone Application
The Prisma AIRS Integration for AI Security Exposure Management can be used to import AI security exposures such as model vulnerabilities, model validation findings (automated red teaming alerts), and posture findings (configuration/policy violations) into AI Security Exposure Management and automate workflows for remediation.
Prisma AIRS is a security platform that can identify security weaknesses in AI models and other AI assets by performing scans, red-teaming attacks, and validating configurations and policies set up by organizations on their AI assets and reporting its violations.
This integration with AI Security Exposure Management retrieves the following types of records from Prisma AIRS:
- AI vulnerabilities - Vulnerabilities identified in open-source AI models used by an organization.
- AI validation findings - Results of automated red teaming or model validation attack tests run by Prisma AIRS to identify model behavior risks.
- AI posture findings - Results of violations of certain rules set up by the organization around AI inventory, for example, "Bl(a)cklist models from HuggingFace" or
"Model Stored In Approved File Format".
Initial release
Required Now Platform applications:
- AI Security Exposure Management 2.0.1
- A Prisma AIRS Account:
- Active client ID and client secret
- An active tenant and its Tenant Service Group ID (TSG ID)
Supported family releases:
Australia
Zurich