Skip to content
AImpact
IT EN
Safety Intermediate Also known as: AI costituzionale · CAI

Constitutional AI

/constitutional ay-eye/

An approach developed by Anthropic where the model is trained to follow a written set of principles (a 'constitution') instead of just case-by-case human preferences.

ShareLinkedInX

In practice

It is the method behind Claude. Upside: behavior rules are explicit and readable, not hidden in millions of ratings. If you pick a model for the company this clarifies the vendor's policy choices.

Related terms

RLAIF RLHF Alignment

Seen in the wild

4 entries mentioning it
  1. CAIS Dangerous Capabilities Evaluations: the standard framework for measuring dangerous LLM capabilities
    High
  2. Claude arrives: the first serious ChatGPT competitor
    High
  3. Constitutional AI: the model self-corrects without humans in the loop
    Medium
  4. Anthropic: an AI safety-focused lab is born
    Landmark
← All terms