API improvements
- You now get the author details in the moderation response.
- New author endpoints to update, retrieve, and list authors.
- Update authors with data to improve trust scoring.
Dashboard features
- View Messages in Context: You can now see moderation items with surrounding conversation context for better decision-making
- Image & Video Blurring: Added ability to blur sensitive images and videos in content view for safer moderation
- Global Queue Option: New global queue setting available. This will remove content from queues if it has been reviewed in any queue.
- Use variables in masking: You can now use $ variables in masking. For example, if you want to wrap matched words in
<b>
tags, you can use<b>$</b>
as the masking value.
Bug fixes and improvements
- Improved dashboard performance and stability
New Author Dashboard
We’ve introduced a new comprehensive author-level dashboard system to enhance end user management.- New Author Detail Pages: Comprehensive user profiles with complete activity
overview, trust levels, moderation history, and detailed analytics
- Summary Tab: Trust level management, sentiment analysis, message statistics, and top violation categories
- Timeline Tab: Complete chronological view of user activity with context grouping and advanced filtering
- Fraud Detection Tab: Advanced behavioral analysis including bot detection, location anomalies, device fingerprinting, and risk assessment
- Metadata Tab: Technical user information and account details
Trust Level System
A new trust level system allows you to tailor moderation actions based on user behavior and find bad actors more easily.- Automated Trust Adjustments: System automatically adjusts trust levels based on user behavior and moderation history
- Manually upgrade or downgrade trust levels as needed
- Trust Level Impact: Trust levels can be consumed via the API or to influence moderation decisions, such as auto-approving content from high-trust users
Author-Level Moderation Actions
- Comprehensive User Actions: Block, suspend, and unblock users directly from their profile pages
- Duration-Based Suspensions: Flexible temporary blocks with customizable time periods (1 hour to 1 year)
- Remember to set up webhooks to handle actions taken on authors