# Constitutional AI

> Constitutional AI

- Skill: `coowoolf/constitutional-ai` (Agent Skill)
- Install (CLI): `npx skillmds@latest add coowoolf/constitutional-ai`
- Raw SKILL.md: https://api.skillmd.com/api/skills/coowoolf/constitutional-ai/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Marketing & Growth
- Author: Coowoolf (https://skillmd.com/u/coowoolf)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/coowoolf/constitutional-ai

---


# Constitutional AI

> Constitutional AI（宪法 AI）

## 概述 / Overview

一种 AI 模型对齐方法，通过 AI 反馈（RLAIF）训练模型遵循一套自然语言原则（即“宪法”），而非单纯依赖人工标注。

## 来源 / Source

- **嘉宾**: Benjamin Mann
- **职位**: Co-founder
- **公司**: Anthropic

## 核心步骤 / Core Steps

1. Define Constitution (Values)
2. Model Generates Response
3. Model Self-Critiques against Constitution
4. Model Rewrites Response
5. Fine-Tune on Revised Data

## 核心原则 / Core Principles

- Define Principles: Establish a constitution of values (e.g., helpful, harmless, honest, human rights).
- Generate & Critique: The model generates a response, then critiques itself based on the constitution.
- Recursive Revision: If the response violates principles, the model rewrites it.
- Supervised Learning: The model is fine-tuned on these revised, compliant outputs.

## 适用场景 / When to Use

在训练大语言模型 (LLM) 时，确保其遵循复杂的人类价值观与安全准则。

## 常见错误 / Common Mistakes

过度依赖简单的用户反馈（RLHF）会导致模型产生讨好倾向；未能定义明确的价值准则。

## 实战案例 / Real-World Example

Anthropic 采用该方法训练 Claude，其中融合了《联合国人权宣言》及其他来源的原则。

## 金句 / Quote

> "First we figure out which ones might apply... then we ask the model itself to critique itself and rewrite its own response in light of the principle."