Text moderation

Text Moderation Overview

We are currently using third-party AI text moderation tools that help us to detect different kinds of undesirable content, including but not limited to: sexual, hate, violence, bullying, promotions, and links to external sites. The AI text moderation tools support over 15 languages and the text models are also trained to understand the semantic meaning of different emojis.

Text Classification

Classes are ordered by severity ranging from level 3 (most severe) to level 0 (benign).

Sexual

 
Hate

 
Violence

 
Bullying

 
Drugs

 
Child Exploitation