Principles

Fallibility and Corrigibility

Both humans and AI can be mistaken; reliable deliberation requires an open, reasoned process of correction.