LESSWRONG
LW

2022 MIRI Alignment Discussion

Jun 15, 2022 by Rob Bensinger

A collection of MIRI write-ups and conversations about alignment released in 2022, following the Late 2021 MIRI Conversations.

317Six Dimensions of Operational Adequacy in AGI Projects

Eliezer Yudkowsky

3y

66

946AGI Ruin: A List of Lethalities

Eliezer Yudkowsky

3y

711

273A central AI alignment problem: capabilities generalization, and the sharp left turn

3y

55

314On how various plans miss the hard bits of the alignment challenge

3y

89

173The inordinately slow spread of good AGI conversations in ML

3y

62

197A note about differential technological development

3y

33

136Brainstorm of things that could force an AI team to burn their lead

3y

8

177AGI ruin scenarios are likely (and disjunctive)

3y

38

65Where I currently disagree with Ryan Greenblatt’s version of the ELK approach

3y

7

159Why all the fuss about recursive self-improvement?

3y

62

50Humans aren't fitness maximizers

3y

46

130Warning Shots Probably Wouldn't Change The Picture Much

3y

42

74What does it mean for an AGI to be 'safe'?

3y

29

136Don't leave your fingerprints on the future

3y

48

134Niceness is unnatural

3y

20

105Contra shard theory, in the context of the diamond maximizer problem

3y

19

64Notes on "Can you control the past"

3y

41

171Decision theory does not imply that we get to have nice things

3y

73

132Superintelligent AI is necessary for an amazing future, but far from sufficient

3y

48

111How could we know that an AGI system will have good consequences?

3y

25

72Distinguishing test from training

3y

11

302A challenge for AGI organizations, and a challenge for readers

Rob Bensinger, Eliezer Yudkowsky

3y

33

102Thoughts on AGI organizations and capabilities work

Rob Bensinger, So8res

3y

17