# Content Moderation

<div data-full-width="true"><figure><img src="https://1182587842-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHyzvhgDL3TrE6D5Hun93%2Fuploads%2FoVnOdBaD8CYD7Y8Dmvzw%2Faivoice_content_moderation.png?alt=media&#x26;token=8954fd86-f027-434e-83cd-a69324de99db" alt=""><figcaption></figcaption></figure></div>

### Moderation Settings

* **Input Content Moderation**

  Scan incoming messages for harmful content before processing

  Output Content Moderation

  Scan AI responses for harmful content before delivery
* **PII Redaction**

  Automatically strip personal data (SSNs, credit cards, emails, etc.)
* **Prompt Injection Scanning**

  Detect and wrap attempts to override AI safety instructions
* **Moderation Sensitivity**

  Minimum character length before input moderation activates (80 chars)

<div data-full-width="true"><figure><img src="https://1182587842-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHyzvhgDL3TrE6D5Hun93%2Fuploads%2FF55iWXoBbRHbBJPvgSOb%2Faivoiceplus_moderation_where.png?alt=media&#x26;token=97392b2f-56f1-44f5-a602-75b363630065" alt=""><figcaption></figcaption></figure></div>