Model Armor screens the prompts and responses for your AI applications. The monitoring dashboard provides you with the data, metrics, and visualizations generated from this screening process. Use these insights to understand how Model Armor protects your AI applications and identifies attempted prompt injections or other malicious activities.
Before you begin
Required role
To get the permissions that
you need to access the monitoring dashboard,
ask your administrator to grant you the
Monitoring Viewer (roles/monitoring.viewer)
IAM role on the project.
For more information about granting roles, see Manage access to projects, folders, and organizations.
This predefined role contains the permissions required to access the monitoring dashboard. To see the exact permissions that are required, expand the Required permissions section:
Required permissions
The following permissions are required to access the monitoring dashboard:
-
monitoring.monitoredResourceDescriptors.list -
monitoring.metricDescriptors.list
You might also be able to get these permissions with custom roles or other predefined roles.
View the monitoring dashboard
In the Google Cloud console, go to the Model Armor page.
Verify that you are viewing the project that you activated Model Armor on.
Go to the Monitoring tab.
View the number of flagged or blocked interactions
On the monitoring dashboard, view the number of interactions, including flagged and blocked interactions. The monitoring dashboard displays the following metrics:
- Total interactions: The total number of prompts and responses that are analyzed by Model Armor.
- Interactions flagged: The number of interactions that violated a configured policy in your Model Armor template or floor settings.
- Interactions blocked: The number of interactions blocked if you
configured Model Armor in the
INSPECT_AND_BLOCKmode. These blocked interactions violated floor settings or templates.
Monitor content security violations
In the Violations over time chart, monitor the number of detected violations over time.
The violations detected are categorized into the following areas:
- All detectors: The total number of violations detected by all detectors, including prompt injections and jailbreaks, malicious URLs, responsible AI, and sensitive data.
- Responsible AI: Content violations detected by safety filters, such as harassment and hate speech. For a complete list of responsible AI categories, see Responsible AI safety filter.
- Sensitive data: Content violations involving the presence of sensitive information types or custom information types that you define. For more information, see Sensitive Data Protection.
For more information about these detectors, see Model Armor filters.
Apply filters to views
When you apply these filters on the monitoring dashboard, all the views on the monitoring dashboard are filtered based on the criteria that you select.
- Templates/Floor settings: Filter views based on the associated template or floor settings.
- Integration points: Filter views based on Model Armor integrations.
- Locations: Filter views based on where an interaction was processed.
- Input types: Filter views based on the type of input evaluated: prompt or response.
Inspect related logs
To explore log entries related to the data on the monitoring dashboard, enable logging in your templates and floor settings. You must have the required IAM roles to view logs.
To inspect the logs for the selected period, on the monitoring dashboard, click More chart options > Inspect related logs.
Download violations data to a PNG or CSV file
To download violations data to a PNG or CSV file, follow these steps:
- In the Violations over time view on the monitoring dashboard, select the period for which you want to download data.
- Click More chart options > Download.
- Click Download PNG or Download CSV to download the data in your preferred format.
What's next
- Learn about Model Armor audit logging.
- Configure logging for Model Armor
- Troubleshoot Model Armor issues.