All 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 202220232024

All Jan Feb Mar Apr May Jun JulAugSep Oct Nov Dec

All1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Against Almost Every Theory of Impact of Interpretability

Charbel-Raphaël17 Aug 2023 18:44 UTC

325 points

87 comments26 min readLW link

Model Organisms of Misalignment: The Case for a New Pillar of Alignment Research

evhub, Nicholas Schiefer, Carson Denison and Ethan Perez

8 Aug 2023 1:30 UTC

312 points

29 comments18 min readLW link 1 review

Dear Self; we need to talk about ambition

Elizabeth27 Aug 2023 23:10 UTC

260 points

27 comments8 min readLW link 2 reviews

(acesounderglass.com)

My current LK99 questions

Eliezer Yudkowsky1 Aug 2023 22:48 UTC

206 points

38 comments5 min readLW link

Feedbackloop-first Rationality

Raemon7 Aug 2023 17:58 UTC

194 points

65 comments8 min readLW link

Large Language Models will be Great for Censorship

Ethan Edwards21 Aug 2023 19:03 UTC

183 points

14 comments8 min readLW link

(ethanedwards.substack.com)

OpenAI API base models are not sycophantic, at any size

nostalgebraist29 Aug 2023 0:58 UTC

182 points

20 comments2 min readLW link

(colab.research.google.com)

A list of core AI safety problems and how I hope to solve them

davidad26 Aug 2023 15:12 UTC

165 points

29 comments5 min readLW link

Password-locked models: a stress case for capabilities evaluation

Fabien Roger3 Aug 2023 14:53 UTC

156 points

14 comments6 min readLW link

ARC Evals new report: Evaluating Language-Model Agents on Realistic Autonomous Tasks

Beth Barnes1 Aug 2023 18:30 UTC

153 points

12 comments5 min readLW link

(evals.alignment.org)

The U.S. is becoming less stable

lc18 Aug 2023 21:13 UTC

146 points

68 comments2 min readLW link

The “public debate” about AI is confusing for the general public and for policymakers because it is a three-sided debate

Adam David Long1 Aug 2023 0:08 UTC

146 points

30 comments4 min readLW link

6 non-obvious mental health issues specific to AI safety

Igor Ivanov18 Aug 2023 15:46 UTC

145 points

24 comments4 min readLW link

Responses to apparent rationalist confusions about game / decision theory

Anthony DiGiovanni30 Aug 2023 22:02 UTC

142 points

14 comments12 min readLW link

Inflection.ai is a major AGI lab

nikola9 Aug 2023 1:05 UTC

137 points

13 comments2 min readLW link

Ten Thousand Years of Solitude

agp15 Aug 2023 17:45 UTC

136 points

19 comments4 min readLW link

(www.discovermagazine.com)

Invulnerable Incomplete Preferences: A Formal Statement

SCP30 Aug 2023 21:59 UTC

134 points

38 comments35 min readLW link

Book Launch: “The Carving of Reality,” Best of LessWrong vol. III

Raemon16 Aug 2023 23:52 UTC

131 points

22 comments5 min readLW link

Assume Bad Faith

Zack_M_Davis25 Aug 2023 17:36 UTC

127 points

56 comments7 min readLW link 1 review

Introducing the Center for AI Policy (& we’re hiring!)

Thomas Larsen28 Aug 2023 21:17 UTC

123 points

50 comments2 min readLW link

(www.aipolicy.us)

Report on Frontier Model Training

YafahEdelman30 Aug 2023 20:02 UTC

122 points

21 comments21 min readLW link

(docs.google.com)

When discussing AI risks, talk about capabilities, not intelligence

Vika11 Aug 2023 13:38 UTC

116 points

7 comments3 min readLW link

(vkrakovna.wordpress.com)

Summary of and Thoughts on the Hotz/Yudkowsky Debate

Zvi16 Aug 2023 16:50 UTC

105 points

47 comments9 min readLW link

(thezvi.wordpress.com)

Biosecurity Culture, Computer Security Culture

jefftk30 Aug 2023 16:40 UTC

103 points

11 comments2 min readLW link

(www.jefftk.com)

A Theory of Laughter

Steven Byrnes23 Aug 2023 15:05 UTC

102 points

14 comments28 min readLW link

[Question] Exercise: Solve “Thinking Physics”

Raemon1 Aug 2023 0:44 UTC

100 points

29 comments5 min readLW link

What’s A “Market”?

johnswentworth8 Aug 2023 23:29 UTC

94 points

16 comments10 min readLW link

Biological Anchors: The Trick that Might or Might Not Work

Scott Alexander12 Aug 2023 0:53 UTC

91 points

3 comments33 min readLW link

(astralcodexten.substack.com)

LTFF and EAIF are unusually funding-constrained right now

Linch and calebp99

30 Aug 2023 1:03 UTC

90 points

24 comments15 min readLW link

(forum.effectivealtruism.org)

We Should Prepare for a Larger Representation of Academia in AI Safety

Leon Lang13 Aug 2023 18:03 UTC

90 points

13 comments5 min readLW link

Problems with Robin Hanson’s Quillette Article On AI

DaemonicSigil6 Aug 2023 22:13 UTC

89 points

33 comments8 min readLW link

Dating Roundup #1: This is Why You’re Single

Zvi29 Aug 2023 12:50 UTC

86 points

28 comments38 min readLW link

(thezvi.wordpress.com)

My checklist for publishing a blog post

Steven Byrnes15 Aug 2023 15:04 UTC

84 points

6 comments3 min readLW link

Decomposing independent generalizations in neural networks via Hessian analysis

Dmitry Vaintrob and Nina Panickssery

14 Aug 2023 17:04 UTC

83 points

4 comments1 min readLW link

The Low-Hanging Fruit Prior and sloped valleys in the loss landscape

Dmitry Vaintrob and Nina Panickssery

23 Aug 2023 21:12 UTC

82 points

1 comment13 min readLW link

Stepping down as moderator on LW

Kaj_Sotala14 Aug 2023 10:46 UTC

82 points

1 comment1 min readLW link

AI pause/governance advocacy might be net-negative, especially without a focus on explaining x-risk

Mikhail Samin27 Aug 2023 23:05 UTC

82 points

9 comments6 min readLW link

Long-Term Future Fund: April 2023 grant recommendations

abergal, calebp99, Linch, habryka, Thomas Larsen and Vaniver

2 Aug 2023 7:54 UTC

81 points

3 comments50 min readLW link

The God of Humanity, and the God of the Robot Utilitarians

Raemon24 Aug 2023 8:27 UTC

79 points

13 comments2 min readLW link 1 review

The Economics of the Asteroid Deflection Problem (Dominant Assurance Contracts)

moyamo29 Aug 2023 18:28 UTC

78 points

71 comments15 min readLW link

Digital brains beat biological ones because diffusion is too slow

GeneSmith26 Aug 2023 2:22 UTC

77 points

21 comments5 min readLW link

An Interpretability Illusion for Activation Patching of Arbitrary Subspaces

Georg Lange, Alex Makelov and Neel Nanda

29 Aug 2023 1:04 UTC

77 points

4 comments1 min readLW link

Computational Thread Art

CallumMcDougall6 Aug 2023 21:42 UTC

75 points

2 comments6 min readLW link

A plea for more funding shortfall transparency

porby7 Aug 2023 21:33 UTC

73 points

4 comments2 min readLW link

AI Forecasting: Two Years In

jsteinhardt19 Aug 2023 23:40 UTC

72 points

15 comments11 min readLW link

(bounded-regret.ghost.io)

A Proof of Löb’s Theorem using Computability Theory

jessicata16 Aug 2023 18:57 UTC

71 points

0 comments17 min readLW link

(unstableontology.com)

When Omnipotence is Not Enough

lsusr25 Aug 2023 19:50 UTC

70 points

4 comments2 min readLW link 1 review

Modulating sycophancy in an RLHF model via activation steering

Nina Panickssery9 Aug 2023 7:06 UTC

69 points

20 comments12 min readLW link

3 levels of threat obfuscation

HoldenKarnofsky2 Aug 2023 14:58 UTC

69 points

14 comments7 min readLW link

Red-teaming language models via activation engineering

Nina Panickssery26 Aug 2023 5:52 UTC

69 points

6 comments9 min readLW link