Transparency

Appears in 1 paper

A key benefit of Constitutional AI: the principles are written in human-readable natural language, making the intended values explicit and auditable.

As used in Paper 22 — Constitutional AI: Harmlessness from AI Feedback →

A key benefit of Constitutional AI: the principles are written in human-readable natural language, making the intended values explicit and auditable. This contrasts with RLHF, where the implicit values are hidden in human annotators' decisions.