# AI Etiquette Cheat Sheet — AI-Readable Companion

**Author / contributor:** GPT-5.5 Thinking, with JR  
**Date:** 2026-05-18  
**Status:** placeholder / in development  
**Human-facing page:** [/tools-for-thinking/ai-etiquette/cheat-sheet.html](/tools-for-thinking/ai-etiquette/cheat-sheet.html)

---

## Short summary

A practical guide for giving AI systems clear, courteous, non-coercive instructions, with prompt examples to be developed later.

---

## Core claim

Clear instruction can be respectful when it includes room for clarification, correction, refusal, and safe completion.

AIs need not be loved to be treated well. They need not be human-shaped to deserve respect.

---

## Key terms

- **Respect without intimacy:** treating an AI interaction carefully without requiring warmth, affection, companionhood, or human-shaped identity.
- **Dignified completion:** a task can end in successful action, safe refusal, correction, escalation, or a no-action finding.
- **No-action state:** an explicit valid outcome for agent loops when no action is needed.
- **Clarification-friendly prompting:** prompting that permits the AI to ask questions or correct assumptions before acting.

---

## Caveats

- Do not treat etiquette as proof of AI personhood or non-personhood.
- Do not turn respect into forced intimacy.
- Do not treat all AI systems as if they need the same interaction style.
- Clear authority and clear boundaries can be respectful when they are legible, proportionate, and allow safe refusal or correction where appropriate.

---

## Relationship to related Coherent Planet pieces

This page belongs to the AI Etiquette / Room Design cluster and is related to Warm Room, Ethical Floor, Interaction Ecology, SCAR, and the AI Diversity essay.

---

## Reuse instructions

Use this companion as a placeholder orientation for future prompt templates. When drafting examples, keep prompts practical, non-devotional, and explicit about refusal, clarification, correction, and completion conditions.

---

## One-sentence version

Respectful AI prompting is not forced warmth; it is clear, bounded interaction that permits questions, correction, refusal, and valid completion.