JustinShovelain

Message

I am the co founder of and researcher at the quantitative long term strategy organization Convergence (see here for our growing list of publications). Over the last sixteen years I have worked with MIRI, CFAR, EA Global, and Founders Fund, and done work in EA...

628

17y

Updating Utility Functions

This post will be about AIs that “refine” their utility function over time, and how it might be possible to construct such systems without giving them undesirable properties. The discussion relates to corrigibility, value learning, and (to a lesser extent) wireheading. We (Joar Skalse and Justin Shovelain) have spent some...

May 9, 202242

Using vector fields to visualise preferences and make them consistent

This post was written for Convergence Analysis by Michael Aird, based on ideas from Justin Shovelain and with ongoing guidance from him. Throughout the post, “I” will refer to Michael, while “we” will refer to Michael and Justin or to Convergence as an organisation. Epistemic status: High confidence in the...

Jan 28, 202043

AI ALIGNMENT FORUM
AF

AI ALIGNMENT FORUM
AF

JustinShovelain

JustinShovelain

JustinShovelain

JustinShovelain

Updating Utility Functions

Using vector fields to visualise preferences and make them consistent

Updating Utility Functions

Using vector fields to visualise preferences and make them consistent