home/categories/machine-learning/orchestra-research-ai-research-skills-07-safety-alignment-constitutional-ai-skill-md
machine-learningdata-ai

constitutional-ai

Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with self-critique/revision, then RLAIF (RL from AI Feedback). Use for safety alignment, reducing harmful outputs without human labels. Powers Claude's safety system.

Orchestra-Research
maintainer
Orchestra-Research
์—…๋ฐ์ดํŠธ๋จ 11/20/2025
์Šคํƒ€
6563
ํฌํฌ
515
quick start

Installation and usage

Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with self-critique/revision, then RLAIF (RL from AI Feedback). Use for safety alignment, reducing harmful outputs without human labels. Powers Claude's safety system.

์„ค์น˜
$ install --globalskills.sh
์‚ฌ์šฉ๋ฒ•

์„ค์น˜ ํ›„ ํ„ฐ๋ฏธ๋„์—์„œ ๋‹ค์Œ ๋ช…๋ น์„ ์‹คํ–‰ํ•˜์—ฌ ์ด ์Šคํ‚ฌ์„ ์‚ฌ์šฉํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค:

skills use constitutional-ai