All 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 202420252026

All Jan Feb Mar Apr May Jun Jul Aug Sep OctNovDec

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

Claude 4.5 Opus’ Soul Document

Richard Weiss28 Nov 2025 23:22 UTC

440 points

44 comments43 min readLW link

Legible vs. Illegible AI Safety Problems

Wei Dai4 Nov 2025 21:39 UTC

370 points

95 comments2 min readLW link

Alignment remains a hard, unsolved problem

evhub27 Nov 2025 8:45 UTC

364 points

96 comments14 min readLW link

Paranoia: A Beginner’s Guide

habryka13 Nov 2025 7:56 UTC

347 points

70 comments13 min readLW link

Why I Transitioned: A Case Study

Fiora Starlight1 Nov 2025 22:58 UTC

324 points

80 comments10 min readLW link

I ate bear fat with honey and salt flakes, to prove a point

aggliu4 Nov 2025 2:00 UTC

324 points

53 comments5 min readLW link

(signoregalilei.com)

Unless its governance changes, Anthropic is untrustworthy

Mikhail Samin29 Nov 2025 5:42 UTC

280 points

68 comments29 min readLW link

(anthropic.ml)

Natural emergent misalignment from reward hacking in production RL

evhub, Monte M, Benjamin Wright and Jonathan Uesato

21 Nov 2025 20:00 UTC

262 points

32 comments9 min readLW link

How Colds Spread

RobertM18 Nov 2025 5:25 UTC

240 points

31 comments10 min readLW link

Why people like your quick bullshit takes better than your high-effort posts

eukaryote28 Nov 2025 20:12 UTC

229 points

27 comments5 min readLW link

(eukaryotewritesblog.com)

You’re always stressed, your mind is always busy, you never have enough time

mingyuan1 Nov 2025 22:07 UTC

226 points

6 comments3 min readLW link

(mingyuan.substack.com)

New Report: An International Agreement to Prevent the Premature Creation of Artificial Superintelligence

peterbarnett, Aaron_Scher, David Abecassis and Brian Abeyta

18 Nov 2025 19:09 UTC

218 points

23 comments3 min readLW link

The Missing Genre: Heroic Parenthood—You can have kids and still punch the sun

Shoshannah Tekofsky29 Nov 2025 1:15 UTC

212 points

27 comments2 min readLW link

(shoshanigans.substack.com)

The Unreasonable Effectiveness of Fiction

Raelifin3 Nov 2025 15:35 UTC

210 points

27 comments8 min readLW link

(raelifin.substack.com)

Stop Applying And Get To Work

Pauliina and plex

23 Nov 2025 22:50 UTC

210 points

56 comments2 min readLW link

Unexpected Things that are People

Ben Goldhaber8 Nov 2025 17:12 UTC

208 points

11 comments4 min readLW link

7 Vicious Vices of Rationalists

Ben Pace16 Nov 2025 7:45 UTC

199 points

33 comments5 min readLW link

Lack of Social Grace is a Lack of Skill

Screwtape3 Nov 2025 4:43 UTC

188 points

26 comments6 min readLW link

Where is the Capital? An Overview

johnswentworth16 Nov 2025 23:18 UTC

186 points

19 comments7 min readLW link

Mourning a life without AI

Nikola Jurkovic8 Nov 2025 4:44 UTC

184 points

63 comments6 min readLW link

(nikolajurkovic.substack.com)

Everyone has a plan until they get lied to the face

Screwtape14 Nov 2025 7:22 UTC

175 points

30 comments7 min readLW link

Gemini 3 is Evaluation-Paranoid and Contaminated

Alice Blair20 Nov 2025 21:02 UTC

173 points

42 comments7 min readLW link

Varieties Of Doom

jdp17 Nov 2025 21:36 UTC

167 points

70 comments57 min readLW link

(minihf.com)

What’s up with Anthropic predicting AGI by early 2027?

ryan_greenblatt3 Nov 2025 16:45 UTC

159 points

16 comments20 min readLW link

The Best Lack All Conviction: A Confusing Day in the AI Village

Zack_M_Davis28 Nov 2025 8:09 UTC

157 points

8 comments6 min readLW link

(zackmdavis.net)

Please, Don’t Roll Your Own Metaethics

Wei Dai12 Nov 2025 22:17 UTC

154 points

65 comments2 min readLW link

Publishing academic papers on transformative AI is a nightmare

Jakub Growiec3 Nov 2025 13:04 UTC

147 points

9 comments4 min readLW link

Tell people as early as possible it’s not going to work out

habryka14 Nov 2025 2:21 UTC

147 points

16 comments2 min readLW link

Condensation

abramdemski9 Nov 2025 19:08 UTC

147 points

14 comments16 min readLW link

Re-rolling environment

Raemon1 Nov 2025 21:46 UTC

140 points

2 comments2 min readLW link

Video games are philosophy’s playground

Rachel Shu17 Nov 2025 6:27 UTC

139 points

17 comments15 min readLW link

(blog.rachelshu.com)

Do not hand off what you cannot pick up

habryka12 Nov 2025 6:32 UTC

137 points

23 comments4 min readLW link

Problems I’ve Tried to Legibilize

Wei Dai9 Nov 2025 10:27 UTC

137 points

24 comments2 min readLW link

The Boring Part of Bell Labs

Elizabeth20 Nov 2025 22:40 UTC

130 points

0 comments15 min readLW link

(acesounderglass.com)

Put numbers on stuff, all the time, otherwise scope insensitivity will eat you

habryka16 Nov 2025 3:04 UTC

129 points

3 comments3 min readLW link

Abstract advice to researchers tackling the difficult core problems of AGI alignment

TsviBT22 Nov 2025 0:53 UTC

129 points

10 comments8 min readLW link

Anthropic is (probably) not meeting its RSP security commitments

habryka18 Nov 2025 23:34 UTC

128 points

22 comments5 min readLW link

ARC progress update: Competing with sampling

Eric Neyman18 Nov 2025 17:22 UTC

126 points

11 comments21 min readLW link

Three positive updates I made about technical grantmaking at Coefficient Giving (fka Open Phil)

jake_mendel26 Nov 2025 1:09 UTC

125 points

3 comments6 min readLW link

Aim for single piece flow

habryka18 Nov 2025 5:22 UTC

115 points

21 comments5 min readLW link

Comparative advantage & AI

Simon Lermen3 Nov 2025 21:50 UTC

114 points

28 comments4 min readLW link

You Are Much More Salient To Yourself Than To Everyone Else

johnswentworth28 Nov 2025 3:14 UTC

114 points

9 comments2 min readLW link

People Seem Funny In The Head About Subtle Signals

johnswentworth6 Nov 2025 4:03 UTC

113 points

36 comments5 min readLW link

AI safety undervalues founders

Ryan Kidd16 Nov 2025 1:59 UTC

112 points

73 comments5 min readLW link

NATO is dangerously unaware that its military edge is slipping

Alexander Gietelink Oldenziel24 Nov 2025 11:40 UTC

112 points

67 comments4 min readLW link

I’ll be sad to lose the puzzles

Ruby23 Nov 2025 19:37 UTC

112 points

21 comments2 min readLW link

How I Learned That I Don’t Feel Companionate Love

johnswentworth12 Nov 2025 4:18 UTC

110 points

32 comments4 min readLW link

From Vitalik: Galaxy brain resistance

Gabriel Alfour10 Nov 2025 13:06 UTC

109 points

2 comments1 min readLW link

(vitalik.eth.limo)

Erasmus: Social Engineering at Scale

Martin Sustrik3 Nov 2025 5:20 UTC

109 points

7 comments4 min readLW link

(www.250bpm.com)

The Tale of the Top-Tier Intellect

Eliezer Yudkowsky3 Nov 2025 20:21 UTC

109 points

58 comments35 min readLW link