Constitutional AI
An Anthropic-authored alignment technique where a model is trained to critique and revise its own outputs against a written set of principles ("constitution") rather than pure human preference labels. Cheaper to scale than RLHF because much of the labelling comes from the model itself.
Related terms
- RLHF
Reinforcement Learning from Human Feedback — a training method where human preferences over pairs of outputs are used to train a reward model, which in turn shapes the base LLM via RL (typically PPO or DPO). Central to aligning modern chat assistants.
- Alignment
The general problem of making an AI system behave according to its designers' or operators' intent, especially as capability grows. Practical alignment work spans RLHF, Constitutional AI, red-teaming, evaluation, and interpretability research.
Related on this site
Where this fits
Constitutional AI is part of the AI Fundamentals vocabulary used in the Generative AI Maturity Framework. See the full glossary for the complete set of 149 defined terms, or take the free maturity assessment to see where your organisation stands.