We’ve introduced a new Classification Method indicator to give you more visibility into how moderation decisions are influenced.

🧠 Understand what influenced a classification

A new filter and tag now show whether a moderation classification was applied using:

• Default rules: the baseline detection patterns used by our technology
• Custom rules: adjustments specifically configured for your organization and moderation needs

This makes it easier to understand when a classification has been influenced by rules tailored to your own environment.

🎯 More transparency in moderation decisions

By surfacing the classification method directly in the message page, teams can better interpret moderation results and understand how specific configurations impact detections.

⚡ Built for future contextual signals

This first iteration lays the foundation for additional context-aware indicators in future releases, including post context and industry-specific rules.

This improvement helps make moderation logic more transparent and easier to interpret in day-to-day workflows.

This improvement builds on an existing feature. Learn more about how it works in our Help Center.