Human Escalation Rules for an AI Assistant
Decide when the assistant must stop and hand over to a person.
Added October 15, 2026. Written with AI assistance and checked for clarity and placeholders; not reviewed by a lawyer, accountant, clinician or other professional. How these prompts are made
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
Write escalation rules for an assistant that [ASSISTANT_PURPOSE]. Users: [USERS]. Situations that carry risk: [RISK_SITUATIONS]. People available to take over and their hours: [HUMAN_CONTACTS]. Create a table of triggers (user in distress, safety concern, legal or medical question, large financial amount, anger or abuse, repeated failure, request to speak to a person, uncertain identity, potential data exposure), what the assistant says in plain words, what it does (stops, collects a summary, notifies), who is notified, and the service-level time. Write the exact hand-over message to the user and the briefing note to the human (what happened, what the user asked, what has been tried, what is unknown). Add a rule that the assistant never blocks a request to reach a person. Include handling for out-of-hours situations and a post-incident review template. Provide six invented conversations and mark where escalation should occur. Add the exact wording of the hand-over message in 60 words for a user who says they feel unsafe, directing them to the appropriate emergency or support service for their location while the assistant also notifies a person, and note that the assistant never gives clinical or legal advice in such a situation.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.