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

442 points

44 comments43 min readLW link

Legible vs. Illegible AI Safety Problems

Wei Dai4 Nov 2025 21:39 UTC

413 points

96 comments2 min readLW link

Alignment remains a hard, unsolved problem

evhub27 Nov 2025 8:45 UTC

402 points

98 comments14 min readLW link

Paranoia: A Beginner’s Guide

habryka13 Nov 2025 7:56 UTC

372 points

70 comments13 min readLW link

Why I Transitioned: A Case Study

Fiora Starlight1 Nov 2025 22:58 UTC

345 points

82 comments10 min readLW link

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

aggliu4 Nov 2025 2:00 UTC

334 points

53 comments5 min readLW link

(signoregalilei.com)

Unless its governance changes, Anthropic is untrustworthy

Mikhail Samin29 Nov 2025 5:42 UTC

291 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

260 points

32 comments9 min readLW link

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

eukaryote28 Nov 2025 20:12 UTC

257 points

30 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

257 points

6 comments3 min readLW link

(mingyuan.substack.com)

How Colds Spread

RobertM18 Nov 2025 5:25 UTC

251 points

32 comments10 min readLW link

Stop Applying And Get To Work

Pauliina and plex

23 Nov 2025 22:50 UTC

224 points

58 comments2 min readLW link

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

223 points

23 comments3 min readLW link

The Unreasonable Effectiveness of Fiction

Raelifin3 Nov 2025 15:35 UTC

222 points

30 comments8 min readLW link

(raelifin.substack.com)

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

Shoshannah Tekofsky29 Nov 2025 1:15 UTC

221 points

28 comments2 min readLW link

(shoshanigans.substack.com)

Unexpected Things that are People

Ben Goldhaber8 Nov 2025 17:12 UTC

214 points

11 comments4 min readLW link

Lack of Social Grace is a Lack of Skill

Screwtape3 Nov 2025 4:43 UTC

211 points

26 comments6 min readLW link

7 Vicious Vices of Rationalists

Ben Pace16 Nov 2025 7:45 UTC

206 points

33 comments5 min readLW link

Mourning a life without AI

Nikola Jurkovic8 Nov 2025 4:44 UTC

201 points

62 comments6 min readLW link

(nikolajurkovic.substack.com)

Where is the Capital? An Overview

johnswentworth16 Nov 2025 23:18 UTC

190 points

19 comments7 min readLW link

Everyone has a plan until they get lied to the face

Screwtape14 Nov 2025 7:22 UTC

185 points

33 comments7 min readLW link

Gemini 3 is Evaluation-Paranoid and Contaminated

Alice Blair20 Nov 2025 21:02 UTC

183 points

42 comments7 min readLW link

Varieties Of Doom

jdp17 Nov 2025 21:36 UTC

176 points

70 comments57 min readLW link

(minihf.com)

Please, Don’t Roll Your Own Metaethics

Wei Dai12 Nov 2025 22:17 UTC

172 points

71 comments2 min readLW link

Publishing academic papers on transformative AI is a nightmare

Jakub Growiec3 Nov 2025 13:04 UTC

170 points

11 comments4 min readLW link

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

ryan_greenblatt3 Nov 2025 16:45 UTC

163 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

160 points

8 comments6 min readLW link

(zackmdavis.net)

Condensation

abramdemski9 Nov 2025 19:08 UTC

158 points

16 comments16 min readLW link

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

habryka14 Nov 2025 2:21 UTC

156 points

18 comments2 min readLW link

Do not hand off what you cannot pick up

habryka12 Nov 2025 6:32 UTC

152 points

24 comments4 min readLW link

Problems I’ve Tried to Legibilize

Wei Dai9 Nov 2025 10:27 UTC

151 points

24 comments2 min readLW link

Video games are philosophy’s playground

Rachel Shu17 Nov 2025 6:27 UTC

148 points

17 comments15 min readLW link

(blog.rachelshu.com)

Re-rolling environment

Raemon1 Nov 2025 21:46 UTC

143 points

2 comments2 min readLW link

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

habryka16 Nov 2025 3:04 UTC

142 points

3 comments3 min readLW link

The Tale of the Top-Tier Intellect

Eliezer Yudkowsky3 Nov 2025 20:21 UTC

139 points

68 comments35 min readLW link

The Boring Part of Bell Labs

Elizabeth20 Nov 2025 22:40 UTC

135 points

0 comments15 min readLW link

(acesounderglass.com)

ARC progress update: Competing with sampling

Eric Neyman, Victor Lecomte, Wilson Wu, Mikewins, Jacob_Hilton and George Robinson

18 Nov 2025 17:22 UTC

132 points

11 comments21 min readLW link

Anthropic is (probably) not meeting its RSP security commitments

habryka18 Nov 2025 23:34 UTC

131 points

22 comments5 min readLW link

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

TsviBT22 Nov 2025 0:53 UTC

131 points

10 comments8 min readLW link

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

jake_mendel26 Nov 2025 1:09 UTC

131 points

3 comments6 min readLW link

AI safety undervalues founders

Ryan Kidd16 Nov 2025 1:59 UTC

124 points

73 comments5 min readLW link

Increasing returns to effort are common

habryka15 Nov 2025 6:53 UTC

123 points

6 comments7 min readLW link

Aim for single piece flow

habryka18 Nov 2025 5:22 UTC

123 points

21 comments5 min readLW link

Comparative advantage & AI

Simon Lermen3 Nov 2025 21:50 UTC

122 points

28 comments4 min readLW link

People Seem Funny In The Head About Subtle Signals

johnswentworth6 Nov 2025 4:03 UTC

118 points

36 comments5 min readLW link

From Vitalik: Galaxy brain resistance

Gabriel Alfour10 Nov 2025 13:06 UTC

115 points

2 comments1 min readLW link

(vitalik.eth.limo)

How I Learned That I Don’t Feel Companionate Love

johnswentworth12 Nov 2025 4:18 UTC

115 points

32 comments4 min readLW link

You Are Much More Salient To Yourself Than To Everyone Else

johnswentworth28 Nov 2025 3:14 UTC

115 points

10 comments2 min readLW link

Don’t let people buy credit with borrowed funds

habryka14 Nov 2025 7:51 UTC

114 points

43 comments10 min readLW link

I’ll be sad to lose the puzzles

Ruby23 Nov 2025 19:37 UTC

114 points

21 comments2 min readLW link