Skip to content

project

Golden Handcuffs

AI safety agenda in which a universal agent delegates control to a human mentor whenever it reaches situations its history has not covered, stated as a guarantee over low-complexity stopping predicates.

Current clusters