x

LESSWRONG

LW

Gunnar_Zarncke's Shortform

by Gunnar_Zarncke

2nd Jan 2021

1 min read

8

This is a special post for quick takes by Gunnar_Zarncke. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page.

Rendering 0/186 comments, sorted by

(show all) Click to highlight new comments since: Today at 6:02 AM

More from Gunnar_Zarncke

Curated and popular this week

186

Gunnar_Zarncke's Shortform — LessWrong

Mentioned in

34Unexpected Conscious Entities

28Consciousness as a Self-Referential Physical Process

Comment Permalink

Gunnar_Zarncke9mo4-3

AI Alignment is now less about aligning an existing agent (reducing rogue objectives in the LLM) and more about discovering where agency exists beyond a single model.
Finding the agent and its objectives in multi-agent, maybe partially human, partially institutional structures.