How to identify missing knowledge in AI Employee responses
Summary
Review conversation logs regularly to find instances where the AI triggers a fallback response, hallucinates an answer, or fails to resolve a customer's inquiry.
Who this article is for
This article is for customer support managers and administrators responsible for optimizing the AI Employee.
Overview
Before you begin
- Access the AI conversation logs
- Filter for conversations that resulted in human handoff
- Look for repeated fallback responses
Step 1: Step 1: Access Conversation Logs
Step 2: Step 2: Look for Fallback Triggers
Step 3: Step 3: Analyze Human Handoffs
Step 4: Step 4: Identify 'Hallucinations'
Step 5: Step 5: Create New Knowledge Entries
Configuration note
We do not guarantee AI accuracy, lead conversion, bookings, customer satisfaction, response quality, revenue, support resolution, or business outcomes. The AI Employee does not fully replace human staff. AI behaviour depends on knowledge sources, instructions, handoff rules, enabled features, connected channels, permissions, configuration, testing, and ongoing monitoring. AI-generated responses should be reviewed before relying on them for sensitive, urgent, legal, financial, medical, billing, or policy matters. The AI Employee may need human handoff for these situations. Knowledge sources must be reviewed, updated, and tested regularly. Usage charges, feature access, fair use rules, or add-on requirements may apply depending on your account setup. Always test before going live. Contact support if an option is not visible. Treat missing knowledge as a feature request from your customers. If multiple people are asking about a specific topic, it's a signal that your website or knowledge base needs to address it clearly.
Important clarification
Ignoring the conversation logs and assuming the AI is functioning perfectly. Without active monitoring and continuous training, the AI's helpfulness will plateau.
Resolution guidance
The AI has the knowledge, but still triggers the fallback response. Why?
The customer may be using terminology the AI doesn't recognize. Add aliases or rephrase the FAQ in the knowledge base to match the language your customers actually use.
Was this article helpful?
Need account-specific help?
If your question requires account access, billing review, workflow inspection, message delivery diagnosis, AI response review, or private configuration review, submit a support request with examples and screenshots.
