Data redaction means masking or removing sensitive information from your data to protect a user's privacy. Redaction is crucial for Google Cloud compliance and maintaining user trust. Data retention policies dictate how long data is stored. A data retention strategy should balance business needs with regulatory requirements. Use security settings to configure data redaction and retention in Agent Assist.
Security settings
Configure your security settings at the project level. You can configure separate security settings for each project and location. You must specify the security settings to apply for each conversation profile. If you don't specify any security settings in a conversation profile, Agent Assist doesn't apply any redaction for that conversation profile. The default and maximum retention window is 30 days.
For more information on the settings described in the following table, see the RPC Security settings reference.
Security settings | Description |
---|---|
redaction_strategy |
The strategy used for redaction. |
redaction_scope |
The data that redaction is applied to. |
inspect_template |
DLP inspect template name. |
deidentify_template |
DLP deidentify template name. |
retention_window_days |
Number of days that data is retained. The default retention window in Agent Assist is 30 days. |
Step 1: Create your inspect and deidentify templates
- Inspect template: Defines sensitive data requiring redaction.
- Deidentify template: Specifies how to mask or remove sensitive data.
Follow these steps to create these templates.
Open the Google Cloud console, and select your project.
Enter a search for Sensitive data protection, and click the link for the security page.
Click the Configuration tab, then enable the Sensitive Data Protection (DLP) API, and create a template.
Step 2: Create security settings
Follow these steps to create security settings.
Open the unified CCAI (Customer Engagement Suite with Google AI) console and select your project.
Click Create Security Settings.
Enter your security settings configuration, and click Create.
Step 3: Specify security settings
Navigate to the Agent Assist console and select your project.
Click Conversation profiles and select a profile.
Navigate to Security Settings and select the display name of the template you created.