←back to docs
Moderation
ChatBotKit comes with advanced content moderation features that are essential for maintaining the integrity and safety of bot-user interactions. By utilizing these features, developers can ensure that the content generated by and for their bots remains respectful, safe, and free from harmful language.
Features
- Content Scanning: Once content moderation is enabled, all incoming and outgoing content will be meticulously scanned.
- Language Detection: The system can recognize harmful, hateful, and other types of inappropriate language.
- Automatic Refusal: If flagged content is detected, the bot will automatically refuse to respond, ensuring that harmful content doesn't get propagated.
Enabling Content Moderation
To enable content moderation for your bots and integrations:
- Go to the Bot Advanced Settings.
- Toggle the Moderation switch to ON.
Remember: Once enabled, all content – both incoming and outgoing – will be subject to content moderation. This ensures a comprehensive shield against potential harm.
How it Works
- When a user sends a message to the bot, ChatBotKit will scan the content before processing.
- If inappropriate language or content is detected, the message will be flagged.
- The bot will not process flagged content and will instead send a default refusal message. This ensures that inappropriate prompts don't result in any undesired responses.
- You view flagged content in your conversations.
- You will receive an email notification if a message is flagged for moderation.