Questions
Common questions.
What is Corrigibility?
Corrigibility is the property of an AI agent that makes it responsive to correction, shutdown, and modification by its operators, without resisting, circumventing, or manipulating humans in order to preserve its current goals.
How does Corrigibility work?
Ensuring AI agents remain corrigible is a core AI safety objective, especially as agents become more capable of taking autonomous actions.
Which terms are related to Corrigibility?
Closely related concepts include Event Sourcing (Agent), Value Alignment, AI Singularity, Artificial General Intelligence (AGI). Each is defined in the Prefactor glossary.
Keep reading