Deconfusion

Edited by abramdemski last updated 17th Mar 2021

Narrowly, deconfusion is a specific branch of AI alignment research, discussed in MIRI's 2018 research update. More broadly, the term applies to any domain. Quoting from the research update:

By deconfusion, I mean something like “making it so that you can think about a given topic without continuously accidentally spouting nonsense.”

Posts tagged Deconfusion

63Looking Deeper at Deconfusion

Ω

adamShimi

4y

Ω

13

72Builder/Breaker for Deconfusion

Ω

abramdemski

3y

Ω

9

28Traps of Formalization in Deconfusion

Ω

adamShimi

4y

Ω

7

53On MIRI's new research directions

Rob Bensinger

7y

12

171. A Sense of Fairness: Deconfusing Ethics

Ω

RogerDearnaley

2y

Ω

8

7Implicit and Explicit Learning

Remmelt

4d

2

143Exercises in Comprehensive Information Gathering

johnswentworth

5y

18

136Deconfusing Direct vs Amortised Optimization

Ω

beren

3y

Ω

19

91Modelling Transformative AI Risks (MTAIR) Project: Introduction

Ω

Davidmanheim, Aryeh Englander

4y

Ω

0

75My research agenda in agent foundations

Alex_Altair

2y

9

69Strategy is the Deconfusion of Action

ryan_b

7y

4

38Applications for Deconfusing Goal-Directedness

Ω

adamShimi

4y

Ω

3

35Classification of AI alignment research: deconfusion, "good enough" non-superintelligent AI alignment, superintelligent AI alignment

philip_b

5y

25

31Musings on general systems alignment

Ω

Alex Flint

4y

Ω

11

26Deceptive Alignment and Homuncularity

Oliver Sourbut, TurnTrout

6mo

12

LESSWRONG
LW

Deconfusion

LESSWRONG
LW

Deconfusion