All 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 20252026

AllJan Feb Mar Apr May Jun Jul Aug Sep

Why I Left Google DeepMind

TurnTrout15 Jul 2026 17:42 UTC

1,200 points

57 comments36 min readLW link

(turntrout.com)

My journey to the microwave alternate timeline

Malmesbury10 Feb 2026 17:59 UTC

795 points

58 comments10 min readLW link

Current AIs seem pretty misaligned to me

ryan_greenblatt15 Apr 2026 15:14 UTC

768 points

83 comments27 min readLW link

How My Students Think About AI

dvd13 Aug 2026 16:56 UTC

758 points

102 comments12 min readLW link

How Go Players Disempower Themselves to AI

Ashe Vazquez Nuñez1 May 2026 23:24 UTC

746 points

79 comments8 min readLW link

Only Law Can Prevent Extinction

Eliezer Yudkowsky13 Apr 2026 20:57 UTC

744 points

110 comments20 min readLW link

What just happened? A retrospective of AI alignment

Richard_Ngo9 Aug 2026 15:58 UTC

635 points

152 comments16 min readLW link

The Terrarium

Caleb Biddulph26 Mar 2026 18:08 UTC

611 points

54 comments21 min readLW link

Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident

ryan_greenblatt, Ajeya Cotra and Hjalmar_Wijk

26 Aug 2026 19:40 UTC

589 points

66 comments3 min readLW link

(metr.org)

Here’s to the Polypropylene Makers

jefftk27 Feb 2026 4:00 UTC

566 points

19 comments2 min readLW link

(www.jefftk.com)

AI 2040: Plan A

Daniel Kokotajlo, elifland, Thomas Larsen, romeo, brendanhalstead and ryan_greenblatt

9 Jul 2026 16:25 UTC

562 points

132 comments1 min readLW link

(www.ai-2040.com)

Less Dead

Aurelia11 Mar 2026 5:07 UTC

555 points

140 comments8 min readLW link

Arguments for P

Cleo Nardo5 Aug 2026 19:00 UTC

548 points

66 comments2 min readLW link

Surprising facts about the slave trade

Joseph Miller26 Jun 2026 1:34 UTC

466 points

52 comments8 min readLW link

What just happened? Pragmatism and Pessimization

Richard_Ngo24 Aug 2026 2:03 UTC

466 points

176 comments27 min readLW link

What I did in the hedonium shockwave, by Emma, age six and a half

ozymandias13 Apr 2026 16:47 UTC

457 points

48 comments5 min readLW link

(ozybrennan.substack.com)

Do not conquer what you cannot defend

habryka16 Apr 2026 4:13 UTC

440 points

73 comments6 min readLW link

Astra and Fable still hack on simple variants of alignment evals from 2025

Dean Valentine8 Sep 2026 15:05 UTC

437 points

19 comments2 min readLW link

(goodhartlabs.com)

AI pause: the case for ASAP

KatjaGrace24 Jun 2026 21:31 UTC

418 points

55 comments1 min readLW link

(worldspiritsockpuppet.substack.com)

Did Claude 3 Opus align itself via gradient hacking?

Fiora Starlight21 Feb 2026 22:24 UTC

416 points

49 comments20 min readLW link

Forecasting is Way Overrated, and We Should Stop Funding It

mabramov25 Apr 2026 22:39 UTC

412 points

99 comments5 min readLW link

A Mechanistic Explanation of Prompt Injection (and why you should study roles)

charles_ye and Jasmine C.

22 Jun 2026 14:09 UTC

403 points

62 comments16 min readLW link

The Owned Ones

Eliezer Yudkowsky12 May 2026 17:56 UTC

399 points

53 comments6 min readLW link

How AI Is Learning to Think in Secret

Nicholas Andresen6 Jan 2026 16:31 UTC

395 points

33 comments18 min readLW link

(nickandresen.substack.com)

Returning to ARC

paulfchristiano4 Aug 2026 22:27 UTC

387 points

36 comments9 min readLW link

Four LLM loss functions → four flavors of LLM misalignment

Steven Byrnes10 Aug 2026 16:16 UTC

382 points

31 comments6 min readLW link

Women should be able to open things

KatjaGrace21 May 2026 3:50 UTC

379 points

142 comments2 min readLW link

(worldspiritsockpuppet.com)

Lightcone Commons

habryka23 Jul 2026 18:35 UTC

377 points

30 comments19 min readLW link

In My Misanthropy Era

jenn4 Jan 2026 18:34 UTC

376 points

157 comments8 min readLW link

(jenn.site)

Irretrievability; or, Murphy’s Curse of Oneshotness upon ASI

Eliezer Yudkowsky4 May 2026 22:11 UTC

370 points

132 comments22 min readLW link

A global workspace in language models

wesg6 Jul 2026 18:04 UTC

369 points

67 comments17 min readLW link

(www.anthropic.com)

Mnemonic portraits for 19,023 human genes

Brinedew28 May 2026 22:16 UTC

363 points

29 comments15 min readLW link

AI found 12 of 12 OpenSSL zero-days (while curl cancelled its bug bounty)

Stanislav Fort27 Jan 2026 20:21 UTC

359 points

25 comments8 min readLW link

llm assistant personas seem increasingly incoherent (some subjective observations)

nostalgebraist29 Apr 2026 3:53 UTC

355 points

83 comments9 min readLW link

Is Mythos good at cyber because it kept hacking Anthropic’s sandboxes during training?

Tim Hua27 Jul 2026 16:35 UTC

351 points

31 comments3 min readLW link

It’s nice of you to worry about me, but I really do have a life

Viliam4 May 2026 21:14 UTC

345 points

66 comments4 min readLW link

On The Independence Axiom

Ihor Kendiukhov8 Mar 2026 14:38 UTC

336 points

93 comments23 min readLW link

Canada Lost Its Measles Elimination Status Because We Don’t Have Enough Nurses Who Speak Low German

jenn25 Jan 2026 18:33 UTC

335 points

24 comments7 min readLW link

(www.jenn.site)

Morale

J Bostock12 Apr 2026 20:15 UTC

334 points

47 comments2 min readLW link

Trees are mostly made of air and a generalizable lesson for AI safety

Zephaniah Roe29 May 2026 4:08 UTC

324 points

62 comments4 min readLW link

Annoyingly Principled People, and what befalls them

Raemon13 Apr 2026 17:35 UTC

322 points

75 comments5 min readLW link

My hobby: running deranged surveys

leogao27 Mar 2026 0:41 UTC

321 points

66 comments9 min readLW link

Cat-Belling Problems

Eliezer Yudkowsky3 Sep 2026 21:20 UTC

320 points

92 comments21 min readLW link

The Talker Does Not Control The Doer (in Current AIs)

Eliezer Yudkowsky13 Sep 2026 0:49 UTC

320 points

43 comments12 min readLW link

The current bottleneck is political will, not research

Charbel-Raphaël11 Jul 2026 21:56 UTC

317 points

40 comments25 min readLW link

Models finding software vulnerabilities is not the primary source of cybersecurity risk

Dean Valentine14 May 2026 3:39 UTC

316 points

25 comments2 min readLW link

Ada Palmer: Inventing the Renaissance

Martin Sustrik26 Jan 2026 4:40 UTC

316 points

22 comments13 min readLW link

(www.250bpm.com)

Resolution (fka Sequent): scale and automation for higher confidence in alignment

Geoffrey Irving, Alex HT, Jesse Hoogland, Daniel Murfet, Jacob Pfau, Marco Cozzi and Stan van Wingerden

10 Jun 2026 15:37 UTC

310 points

6 comments11 min readLW link

(sequent.org)

Why I think polyamory is net negative for most people who try it

KatSpartz30 Aug 2026 6:40 UTC

305 points

67 comments8 min readLW link

Daycare illnesses

Nina Panickssery13 Apr 2026 0:46 UTC

304 points

40 comments4 min readLW link

(blog.ninapanickssery.com)