AI ALIGNMENT FORUM
AF

Nominated Posts for the 2019 Review

Posts need at least 2 nominations to continue into the Review Phase.

Nominate posts that you have personally found useful and important.

Sort by: fewest nominations

51The Commitment Races problem

Daniel Kokotajlo

5y

36

1 • 0

20How common is it for one entity to have a 3+ year technological lead on its nearest competitor?

Daniel Kokotajlo

5y

1

1 • 0

27AGI will drastically increase economies of scale

5y

15

1 • 0

25Misconceptions about continuous takeoff

Matthew Barnett

5y

12

1 • 0

28But exactly how complex and fragile?

5y

30

2 • 1

34The strategy-stealing assumption

Paul Christiano

5y

45

2 • 3

29Reframing Impact

5y

4

2 • 1

53Gradient hacking

5y

33

2 • 2

38Reframing Superintelligence: Comprehensive AI Services as General Intelligence

6y

29

2 • 2

37The Credit Assignment Problem

5y

32

2 • 1

50Thoughts on Human Models

Ramana Kumar, Scott Garrabrant

6y

9

2 • 1

39Soft takeoff can still lead to decisive strategic advantage

Daniel Kokotajlo

5y

34

2 • 4

50AI Safety "Success Stories"

5y

11

2 • 1

55Utility ≠ Reward

Vladimir Mikulik

5y

16

2 • 2

59Selection vs Control

5y

14

2 • 2

55Seeking Power is Often Convergently Instrumental in MDPs

Alex Turner, Logan Riggs Smith

5y

34

2 • 2

103What failure looks like

Paul Christiano

6y

28

2 • 2

27Strategic implications of AIs' ability to coordinate at low cost, for example by merging

6y

22

2 • 1

30Six AI Risk/Strategy Ideas

5y

14

2 • 1

26Classifying specification problems as variants of Goodhart's Law

Victoria Krakovna

5y

5

2 • 1

66Chris Olah’s views on AGI safety

5y

30

2 • 2

63Debate on Instrumental Convergence between LeCun, Russell, Bengio, Zador, and More

5y

17

2 • 2

56Alignment Research Field Guide

6y

6

2 • 2

47Why Subagents?

5y

12

2 • 1

57Evolution of Modularity

5y

6

2 • 1

58Risks from Learned Optimization: Introduction

Evan Hubinger, Chris van Merwijk, Vladimir Mikulik, Joar Skalse, Scott Garrabrant

5y

33

3 • 3

53Understanding “Deep Double Descent”

5y

35

3 • 4

90The Parable of Predict-O-Matic

5y

16

5 • 4

2019 Review Discussion