Paul Colognese

Ex­plain­ing the AI Align­ment Prob­lem to Ti­be­tan Bud­dhist Monks

Ano­ma­lous Con­cept De­tec­tion for De­tect­ing Hid­den Cognition

Hid­den Cog­ni­tion De­tec­tion Meth­ods and Bench­marks

Notes on In­ter­nal Ob­jec­tives in Toy Models of Agents

In­ter­nal Tar­get In­for­ma­tion for AI Oversight

[Question] Po­ten­tial al­ign­ment tar­gets for a sovereign su­per­in­tel­li­gent AI

High-level in­ter­pretabil­ity: de­tect­ing an AI’s objectives

[Linkpost] Fron­tier AI Task­force: first progress report

Aligned AI via mon­i­tor­ing ob­jec­tives in Au­toGPT-like systems

Towards a solu­tion to the al­ign­ment prob­lem via ob­jec­tive de­tec­tion and eval­u­a­tion

De­ci­sion Trans­former Interpretability

Paul Colog­nese’s Shortform

Au­dit­ing games for high-level interpretability

De­cep­tion?! I ain’t got time for that!