Constitutional AI
Aligning a model using written principles rather than human labels alone.
Overview
Constitutional AI has the model critique and revise its own outputs against a written set of principles, then trains on the result. It solves the cost and inconsistency of collecting human judgements for every harmful output: a principle can be stated once and applied to thousands of cases. It exists as an alternative to purely human feedback, with the trade that the principles themselves are authored and therefore carry their authors' judgement.
Technology Graph
How Constitutional AI connects to the rest of the stack.
Applied to
- Alignment technique Safety & Alignment Constitutional AI used for Alignment High ✓ source
Alternatives
Comparisons
Side-by-side breakdowns with the alternatives above.