NSFW Content Blurring in Sprinklr Social
Updated
Overview
Sprinklr uses advanced machine learning models to detect and manage NSFW, triggering, and objectionable content across data ingested from multiple sources. Given the wide range of incoming social media streams, there is always a possibility of sensitive content appearing in the feed. To address this, Sprinklr can automatically blur or censor NSFW media within posts.
In addition, Sprinklr's ML models are trained to identify accounts that frequently share such content and enable further actions based on your configured policies.

Scope of NSFW Content Blurring in Sprinklr Social
Sprinklr Social provides flexible and configurable options to detect, manage, and control NSFW, triggering, and objectionable content across incoming social media streams. Below are the key areas where content blurring and moderation can be applied:
1. Blur or Remove Objectionable Content
You can choose how NSFW or objectionable media is handled—either by blurring it or removing it entirely from view. This flexibility allows organizations to define the level of sensitivity required and apply the most suitable moderation approach for their needs.
2. Account-Level Moderation and Blocking
Certain accounts may repeatedly post vulgar, offensive, or sensitive content to generate engagement. In such cases, you can share these account IDs with the Sprinklr Support team. These accounts can then be blocked entirely or their content can be automatically blurred in your feed. Once flagged, no further posts from these accounts will be displayed as per the configured action.
3. Customized Content Blurring
Organizations may have specific requirements to filter or blur particular objects or themes—for example, removing gun-related violence or ensuring competitor logos do not appear in feeds. Sprinklr’s image detection models can be trained to identify and blur or exclude such custom entities. For specialized use cases not already supported, you can contact the Support team to configure custom detection rules.
4. Account Bio-Based Filtering
Some social media profiles include NSFW links, offensive language, or other objectionable content in their bios. Sprinklr’s machine learning models can detect such keywords and automatically blur or flag these accounts to prevent exposure to potentially sensitive content.
Note: This blurring behavior is managed through the dynamic property (DP) - MARK_MEDIA_AS_NSFW, which determines whether incoming media is automatically treated as sensitive.
When enabled, all media in messages are flagged as NSFW, routed through the sensitive-content workflow, and displayed blurred by default until users choose to reveal it. When disabled, media follows standard processing and remains visible unless it is detected or manually marked as sensitive.
To enable this capability, please contact the Support team at tickets@sprinklr.com.