Dec 12, 2023, 6:14 PM

458 points

206 comments33 min readLW link 2 reviews

Speaking to Congressional staffers about AI risk

Orpheus16 and hath

Dec 4, 2023, 11:08 PM

312 points

25 comments15 min readLW link 1 review

Constellations are Younger than Continents

Jeffrey HeningerDec 19, 2023, 6:12 AM

263 points

21 comments2 min readLW link

AI Control: Improving Safety Despite Intentional Subversion

Buck, Fabien Roger, ryan_greenblatt and Kshitij Sachan

Dec 13, 2023, 3:51 PM

236 points

24 comments10 min readLW link 4 reviews

Thoughts on “AI is easy to control” by Pope & Belrose

Steven ByrnesDec 1, 2023, 5:30 PM

197 points

63 comments14 min readLW link 1 review

Is being sexy for your homies?

ValentineDec 13, 2023, 8:37 PM

193 points

100 comments14 min readLW link 2 reviews

“Humanity vs. AGI” Will Never Look Like “Humanity vs. AGI” to Humanity

Thane RuthenisDec 16, 2023, 8:08 PM

191 points

34 comments5 min readLW link

Effective Aspersions: How the Nonlinear Investigation Went Wrong

TracingWoodgrainsDec 19, 2023, 12:00 PM

188 points

172 comments LW link 2 reviews

re: Yudkowsky on biological materials

bhauthDec 11, 2023, 1:28 PM

182 points

30 comments5 min readLW link

Critical review of Christiano’s disagreements with Yudkowsky

Vanessa KosoyDec 27, 2023, 4:02 PM

176 points

40 comments15 min readLW link

The ‘Neglected Approaches’ Approach: AE Studio’s Alignment Agenda

Cameron Berg, Judd Rosenblatt, AE Studio and Marc Carauleanu

Dec 18, 2023, 8:35 PM

175 points

22 comments12 min readLW link 1 review

2023 Unofficial LessWrong Census/Survey

ScrewtapeDec 2, 2023, 4:41 AM

169 points

81 comments1 min readLW link

How useful is mechanistic interpretability?

ryan_greenblatt, Neel Nanda, Buck and habryka

Dec 1, 2023, 2:54 AM

167 points

54 comments25 min readLW link

The likely first longevity drug is based on sketchy science. This is bad for science and bad for longevity.

BobBurgersDec 12, 2023, 2:42 AM

161 points

34 comments5 min readLW link

Most People Don’t Realize We Have No Idea How Our AIs Work

Thane RuthenisDec 21, 2023, 8:02 PM

159 points

42 comments1 min readLW link

Succession

Richard_NgoDec 20, 2023, 7:25 PM

159 points

48 comments11 min readLW link

(www.narrativeark.xyz)

The Plan − 2023 Version

johnswentworthDec 29, 2023, 11:34 PM

152 points

40 comments31 min readLW link 1 review

Discussion: Challenges with Unsupervised LLM Knowledge Discovery

Seb Farquhar, Vikrant Varma, zac_kenton, gasteigerjo, Vlad Mikulik and Rohin Shah

Dec 18, 2023, 11:58 AM

147 points

21 comments10 min readLW link

AI Views Snapshots

Rob BensingerDec 13, 2023, 12:45 AM

142 points

61 comments1 min readLW link

The Dark Arts

lsusr and Lyrongolem

Dec 19, 2023, 4:41 AM

134 points

49 comments9 min readLW link

Current AIs Provide Nearly No Data Relevant to AGI Alignment

Thane RuthenisDec 15, 2023, 8:16 PM

131 points

157 comments8 min readLW link 1 review

Natural Latents: The Math

johnswentworth and David Lorell

Dec 27, 2023, 7:03 PM

128 points

40 comments12 min readLW link 2 reviews

Deep Forgetting & Unlearning for Safely-Scoped LLMs

scasperDec 5, 2023, 4:48 PM

126 points

30 comments13 min readLW link

Bayesian Injustice

Kevin DorstDec 14, 2023, 3:44 PM

124 points

10 comments6 min readLW link

(kevindorst.substack.com)

The LessWrong 2022 Review

habrykaDec 5, 2023, 4:00 AM

115 points

43 comments4 min readLW link

Mapping the semantic void: Strange goings-on in GPT embedding spaces

mwatkinsDec 14, 2023, 1:10 PM

114 points

31 comments14 min readLW link

What I Would Do If I Were Working On AI Governance

johnswentworthDec 8, 2023, 6:43 AM

110 points

32 comments10 min readLW link

“AI Alignment” is a Dangerously Overloaded Term

RokoDec 15, 2023, 2:34 PM

108 points

100 comments3 min readLW link

[Question] How do you feel about LessWrong these days? [Open feedback thread]

Bird ConceptDec 5, 2023, 8:54 PM

107 points

285 comments1 min readLW link

Fact Finding: Attempting to Reverse-Engineer Factual Recall on the Neuron Level (Post 1)

Neel Nanda, Senthooran Rajamanoharan, János Kramár and Rohin Shah

Dec 23, 2023, 2:44 AM

106 points

10 comments22 min readLW link 2 reviews

On the future of language models

owencbDec 20, 2023, 4:58 PM

105 points

17 comments LW link

The Witness

Richard_NgoDec 3, 2023, 10:27 PM

105 points

5 comments14 min readLW link

(www.narrativeark.xyz)

Nonlinear’s Evidence: Debunking False and Misleading Claims

KatWoodsDec 12, 2023, 1:16 PM

104 points

171 comments LW link

[Valence series] 1. Introduction

Steven ByrnesDec 4, 2023, 3:40 PM

99 points

16 comments16 min readLW link 2 reviews

A Crisper Explanation of Simulacrum Levels

Thane RuthenisDec 23, 2023, 10:13 PM

92 points

13 comments13 min readLW link

Nietzsche’s Morality in Plain English

Arjun PanicksseryDec 4, 2023, 12:57 AM

92 points

14 comments4 min readLW link 1 review

(arjunpanickssery.substack.com)

Meaning & Agency

abramdemskiDec 19, 2023, 10:27 PM

91 points

17 comments14 min readLW link

Prediction Markets aren’t Magic

SimonMDec 21, 2023, 12:54 PM

90 points

29 comments3 min readLW link

Based Beff Jezos and the Accelerationists

ZviDec 6, 2023, 4:00 PM

90 points

29 comments12 min readLW link

(thezvi.wordpress.com)

[Valence series] 2. Valence & Normativity

Steven ByrnesDec 7, 2023, 4:43 PM

88 points

7 comments28 min readLW link 1 review

Some for-profit AI alignment org ideas

Eric HoDec 14, 2023, 2:23 PM

86 points

19 comments9 min readLW link

A Universal Emergent Decomposition of Retrieval Tasks in Language Models

Alexandre Variengien and Eric Winsor

Dec 19, 2023, 11:52 AM

84 points

3 comments10 min readLW link

(arxiv.org)

Refusal mechanisms: initial experiments with Llama-2-7b-chat

Andy Arditi and Oscar Obeso

Dec 8, 2023, 5:08 PM

82 points

7 comments7 min readLW link

Studying The Alien Mind

Quentin FEUILLADE--MONTIXI and NicholasKees

Dec 5, 2023, 5:27 PM

80 points

10 comments15 min readLW link

EU policymakers reach an agreement on the AI Act

tlevinDec 15, 2023, 6:02 AM

78 points

7 comments7 min readLW link

OpenAI: Leaks Confirm the Story

ZviDec 12, 2023, 2:00 PM

77 points

9 comments16 min readLW link

(thezvi.wordpress.com)

Send us example gnarly bugs

Beth Barnes, Megan Kinniment and Tao Lin

Dec 10, 2023, 5:23 AM

77 points

10 comments2 min readLW link

The Offense-Defense Balance Rarely Changes

Maxwell TabarrokDec 9, 2023, 3:21 PM

77 points

23 comments3 min readLW link

(maximumprogress.substack.com)

MATS Summer 2023 Retrospective

utilistrutil, Juan Gil, Ryan Kidd, Christian Smith, McKennaFitzgerald and LauraVaughan

Dec 1, 2023, 11:29 PM

77 points

34 comments26 min readLW link

[Valence series] 3. Valence & Beliefs

Steven ByrnesDec 11, 2023, 8:21 PM

77 points

12 comments21 min readLW link 1 review