---
id: wiki-2026-0508-hhh
title: HHH (Helpful, Harmless, Honest)
category: 10_Wiki/Topics
status: duplicate
canonical_id: wiki-2026-0508-ethics-ai
duplicate_of: "[[Ethics & AI]]"
aliases: [HHH, helpful harmless honest, Anthropic alignment, Constitutional AI]
source_trust_level: A
confidence_score: 0.94
verification_status: redirected
tags: [duplicate, alignment, hhh, anthropic]
last_reinforced: 2026-05-10
github_commit: pending
---

# HHH (Helpful, Harmless, Honest)

> **이 문서는 [[Ethics & AI]] 의 specialization 입니다.** Canonical 문서로 redirect.

## 핵심 요약 (HHH-specific)
- 매 Anthropic 의 origin (Askell 2021).
- **Helpful**: 매 user task 의 enable.
- **Harmless**: 매 harm 의 avoid.
- **Honest**: 매 truthful, 매 hallucinate X.
- 매 trade-off: 매 helpful ↔ harmless 의 conflict (e.g., dual-use).
- 매 Constitutional AI (Bai 2022) 의 building block.

## 🔗 Graph
- 부모: [[Ethics & AI]] (canonical)
- Adjacent: [[Constitutional-AI]] · [[RLHF]]

## 🕓 변경 이력
| 날짜 | 변경 |
|---|---|
| 2026-05-08 | Phase 1 |
| 2026-05-10 | 중복 처리 — canonical 문서로 redirect |