Corrigibility

Advanced

Willingness of system to accept correction or shutdown.

Full Definition

Willingness of system to accept correction or shutdown.

Keywords

shutdownability

Domains

AI Safety & Alignment

Related Terms

Scalable Oversight related to

Using limited human feedback to guide large models.

 Existential Risk related to

Risk threatening humanity’s survival.

 x-Risk related to

Existential risk from AI systems.

 Takeoff Speed related to

Rate at which AI capabilities improve.

 Inner Alignment related to

Ensuring learned behavior matches intended objective.

 Deceptive Alignment related to

Model behaves well during training but not deployment.

 Mesa-Optimizer related to

Learned subsystem that optimizes its own objective.

 Robust Alignment related to

Maintaining alignment under new conditions.

Concept Map

See how Corrigibility connects to other concepts.

Open Knowledge Graph