How to set sensitive-topic rules for your AI Employee

    CategoryAI Employee
    Article typeDocumentation
    Estimated reading time 5 min
    Last reviewedRecently reviewed
    Applies toGEVADE CRM accounts where the feature is enabled
    Access requiredAccess may depend on your user permissions
    Plan/configuration noteFeature availability may depend on your selected plan and account configuration

    Summary

    Define explicit rules in the AI settings instructing it to immediately halt the conversation and trigger human handoff if topics like legal threats, medical advice, or severe complaints arise.

    Who this article is for

    This article is for business owners and administrators managing risk and compliance.

    Overview

    Certain topics require human empathy, legal caution, or specialized expertise that an AI should never attempt to handle. Setting strict rules for sensitive topics protects your business and ensures customers receive appropriate care.

    Before you begin

    • Identify sensitive topics for your specific industry
    • Write explicit instructions forbidding the AI from discussing these topics
    • Configure immediate human handoff for these triggers

    Step 1: Step 1: Identify Your Sensitive Topics

    Make a list of topics the AI must avoid. Common examples include requests for refunds, legal threats, medical questions, discussions of competitors, or abusive language.

    Step 2: Step 2: Access Business Rules

    Navigate to Settings > AI Employee > Business Rules or Custom Instructions.

    Step 3: Step 3: Write Explicit Prohibitions

    Use strong, unequivocal language. Example: 'NEVER provide medical advice. If a user asks about symptoms, immediately state you cannot provide medical information and trigger human handoff.'

    Step 4: Step 4: Define the Escape Phrase

    Give the AI a specific script to use when a sensitive topic is detected. Example: 'For your privacy and safety, I need to transfer this conversation to a senior specialist. Please hold.'

    Step 5: Step 5: Test the Rules

    Use the AI Simulator to ask questions about the sensitive topics. Verify that the AI refuses to answer the question and immediately executes the handoff protocol.

    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. AI safety requires constant vigilance. Review conversations regularly to ensure the AI isn't finding loopholes around your sensitive topic rules.

    Important clarification

    Using vague language like 'Try to avoid legal topics.' You must use absolute commands like 'NEVER discuss legal topics.'

    Resolution guidance

    The AI is still trying to answer sensitive questions.

    Your instructions may be buried too deep in the prompt. Move the sensitive topic rules to the very top of the Custom Instructions so they are prioritized.

    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.