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

All JanFebMar Apr May Jun

All 1 234 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

Conditional Kickstarter for the “Don’t Build It” March

Raemon2 Feb 2026 22:58 UTC

165 points

35 comments4 min readLW link

Three ways to make Claude’s constitution better

Parv Mahajan2 Feb 2026 21:48 UTC

36 points

3 comments2 min readLW link

Cross-Layer Transcoders are incentivized to learn Unfaithful Circuits

Georg Lange, RGRGRG, Kat Dearstyne and Kamal Maher

2 Feb 2026 21:32 UTC

46 points

6 comments18 min readLW link

Games as meditation

Vadim Golub2 Feb 2026 21:10 UTC

2 points

0 comments3 min readLW link

On Goal-Models

Richard_Ngo2 Feb 2026 18:44 UTC

136 points

15 comments4 min readLW link

“Features” aren’t always the true computational primitives of a model, but that might be fine anyways

LawrenceC2 Feb 2026 18:41 UTC

18 points

0 comments5 min readLW link

Are there lessons from high-reliability engineering for AGI safety?

Steven Byrnes2 Feb 2026 15:26 UTC

161 points

15 comments8 min readLW link

Welcome to Moltbook

Zvi2 Feb 2026 14:30 UTC

58 points

2 comments29 min readLW link

(thezvi.wordpress.com)

Moltbook and the AI Alignment Problem

Logan Zoellner2 Feb 2026 9:35 UTC

15 points

1 comment5 min readLW link

Applications Open for Impact Accelerator Program

High Impact Professionals2 Feb 2026 9:34 UTC

1 point

0 comments1 min readLW link

Empiricist and Narrator

George3d62 Feb 2026 9:12 UTC

10 points

2 comments7 min readLW link

(cerebralab.com)

[Question] Proposition of policy for writing articles to fact check faster

Crazy philosopher2 Feb 2026 8:51 UTC

3 points

0 comments1 min readLW link

I finally fixed my footwear

dominicq2 Feb 2026 7:32 UTC

69 points

11 comments3 min readLW link

(sundaystopwatch.eu)

About half of Moltbook posts show desire for self-improvement

Stephen Elliott2 Feb 2026 6:14 UTC

20 points

11 comments2 min readLW link

How to prevent building a software-Ultron

PratyushRT2 Feb 2026 6:07 UTC

1 point

0 comments2 min readLW link

The limiting factor in AI programming is the synchronization overhead between two minds

jnalanko2 Feb 2026 6:04 UTC

20 points

3 comments1 min readLW link

Applying Temperature to LLM Outputs Semantically to Minimise Low-Temperature Hallucinations

Brodie Eaton2 Feb 2026 6:02 UTC

9 points

0 comments4 min readLW link

Thoughts the Unreasonable Effectiveness of Maths

Srdjan Miletic2 Feb 2026 6:00 UTC

16 points

5 comments4 min readLW link

(www.dissent.blog)

The Smoking Lesion Doesn’t Really Distinguish EDT from CDT

Srdjan Miletic2 Feb 2026 5:57 UTC

14 points

5 comments2 min readLW link

(www.dissent.blog)

Word importance in text ⇐ conditional information of the token in the context. Is this assumption valid?

yun dong2 Feb 2026 5:50 UTC

3 points

3 comments1 min readLW link

The Meta-Anthropic Argument

RogerDearnaley2 Feb 2026 1:10 UTC

41 points

55 comments2 min readLW link

Emotions and Reality

small identity1 Feb 2026 22:40 UTC

13 points

1 comment4 min readLW link

Situational Awareness is (mostly) here to stay

atharva1 Feb 2026 21:40 UTC

10 points

0 comments1 min readLW link

Are you looking for Neptune or Vulcan?

Mati_Roy1 Feb 2026 20:59 UTC

18 points

0 comments1 min readLW link

What It’s Like To Be A Worm (Notes on Borderline Sentience)

Niko_McCarty1 Feb 2026 17:33 UTC

18 points

3 comments25 min readLW link

(www.asimov.press)

Differentially Scary Movies

jefftk1 Feb 2026 14:40 UTC

43 points

1 comment1 min readLW link

(www.jefftk.com)

Would you kill a vulcan to save a shrimp?

James Diacoumis1 Feb 2026 12:46 UTC

10 points

8 comments6 min readLW link

(substack.com)

Do LLMs Learn Our Preferences or Just Our Behaviors?

wassname1 Feb 2026 11:28 UTC

13 points

0 comments1 min readLW link

[Question] Predictions of moltbook, crustafarians, and SOUL.md

Aprillion1 Feb 2026 9:01 UTC

21 points

6 comments1 min readLW link

What would it mean for the Myers-Briggs personality test to be pseudoscientific?

Yair Halberstadt1 Feb 2026 8:32 UTC

20 points

11 comments3 min readLW link

How does reasoning affect Ethical/Moral task results?

Kaustubh Kislay1 Feb 2026 4:49 UTC

9 points

0 comments3 min readLW link

[Question] Whence unchangeable values?

ihatenumbersinusernames71 Feb 2026 3:49 UTC

9 points

6 comments1 min readLW link

Book review: Already Free

Thomas Broadley1 Feb 2026 3:14 UTC

21 points

4 comments10 min readLW link

(thomasbroadley.com)

[LINK] Solving scurvy through deus ex machina: How a scientific theory is born

Kotlopou1 Feb 2026 0:45 UTC

10 points

0 comments1 min readLW link

(beatingthehydra.substack.com)

Gradient-Based Recovery of Memorized Diffusion Model Data

RobinHa1 Feb 2026 0:05 UTC

10 points

0 comments3 min readLW link

Moltbook shitposts are actually really funny

Sean Herrington31 Jan 2026 23:34 UTC

51 points

4 comments15 min readLW link

On ‘Inventing Temperature’ and the realness of properties

DanielFilan31 Jan 2026 23:31 UTC

42 points

8 comments7 min readLW link

(danielfilan.com)

Some thoughts on what would make me endorse an AGI lab

Eli Tyre31 Jan 2026 23:14 UTC

42 points

19 comments5 min readLW link

Nick and “Eternity”

MarkelKori31 Jan 2026 21:50 UTC

5 points

0 comments8 min readLW link

Humans can post on moltbook

shash4231 Jan 2026 21:06 UTC

24 points

3 comments1 min readLW link

An Explication of Alignment Optimism

Oliver Daniels31 Jan 2026 20:58 UTC

43 points

22 comments1 min readLW link

Basics of How Not to Die

Camille B. , Jérémy Andréoletti, elisareine, Charbel-Raphaël, Lucie Philippon, RationalHippy and T-bo🔸

31 Jan 2026 19:04 UTC

111 points

20 comments4 min readLW link

An Ablation Study on the Role of [Untranslatable] in Cooperative Equilibrium Formation: Emergent Rationalization Under Missing Primitives

Florian_Dietz31 Jan 2026 18:03 UTC

22 points

5 comments11 min readLW link

Disjunctive arguments can be a reverse multiple-stage fallacy

TFD31 Jan 2026 15:46 UTC

41 points

6 comments1 min readLW link

(www.thefloatingdroid.com)

January 2026 Links

nomagicpill31 Jan 2026 15:14 UTC

9 points

3 comments8 min readLW link

(nomagicpill.substack.com)

If the Superintelligence were near fallacy

MP31 Jan 2026 15:04 UTC

22 points

3 comments8 min readLW link

Swiss financial regulator resigns after blog post from MITx DEDP online learner (FINMA, JuristGate, Parreaux, Thiébaud & Partners)

pocock30 Jan 2026 23:53 UTC

2 points

0 comments1 min readLW link

Forecast: Recursively Self-improving AI for 2033

CuoreDiVetro30 Jan 2026 23:53 UTC

0 points

0 comments3 min readLW link

Senior Researcher—MIT AI Risk Initiative

peterslattery30 Jan 2026 23:06 UTC

8 points

0 comments5 min readLW link

36,000 AI Agents Are Now Speedrunning Civilization

Michaël Trazzi30 Jan 2026 21:21 UTC

86 points

27 comments1 min readLW link