Lex­i­con of Life Regulation

henophilia3 Feb 2026 22:39 UTC
−15 points
0 comments15 min readLW link
(blog.hermesloom.org)

‘In­vent­ing the Re­nais­sance’ Review

Commander Zander3 Feb 2026 22:01 UTC
60 points
2 comments3 min readLW link

Con­crete re­search ideas on AI personas

3 Feb 2026 21:50 UTC
69 points
10 comments6 min readLW link

Progress links and short notes, 2026-01-26

jasoncrawford3 Feb 2026 21:42 UTC
11 points
0 comments5 min readLW link
(rootsofprogress.substack.com)

The Pro­jec­tion Prob­lem: Two Pit­falls in AI Safety Research

Shivam3 Feb 2026 21:03 UTC
6 points
2 comments6 min readLW link

New AI safety fund­ing newsletter

Bryce Robertson3 Feb 2026 20:23 UTC
42 points
0 comments1 min readLW link

dis­gust at util­ity maximization

pantalaimon3 Feb 2026 20:07 UTC
1 point
4 comments1 min readLW link

METR have re­leased Time Hori­zons 1.1

Sean Herrington3 Feb 2026 19:48 UTC
33 points
0 comments1 min readLW link
(metr.org)

AI Safety at the Fron­tier: Paper High­lights of Jan­uary 2026

gasteigerjo3 Feb 2026 18:56 UTC
22 points
0 comments9 min readLW link
(aisafetyfrontier.substack.com)

Un­less That Claw Is The Fa­mous OpenClaw

Zvi3 Feb 2026 15:00 UTC
39 points
5 comments16 min readLW link
(thezvi.wordpress.com)

Ex­po­nen­tial take­off of mediocrity

Valerii K.3 Feb 2026 14:41 UTC
4 points
5 comments32 min readLW link

AI for Hu­man Rea­son­ing for Rationalists

Oliver Sourbut3 Feb 2026 13:22 UTC
29 points
0 comments4 min readLW link
(www.oliversourbut.net)

Con­di­tion­al­iza­tion Con­founds Inoc­u­la­tion Prompt­ing Results

3 Feb 2026 11:50 UTC
78 points
5 comments19 min readLW link

The Atoms of Knowl­edge Aren’t Universal

Jonas Hallgren3 Feb 2026 10:52 UTC
19 points
4 comments13 min readLW link
(equilibria1.substack.com)

What did we learn from the AI Village in 2025?

Shoshannah Tekofsky3 Feb 2026 9:52 UTC
63 points
5 comments10 min readLW link
(theaidigest.org)

Thought Edit­ing: Steer­ing Models by Edit­ing Their Chain of Thought

3 Feb 2026 9:51 UTC
20 points
0 comments5 min readLW link

De­sign in­ter­na­tional AI pro­jects with DAID in mind

wdmacaskill3 Feb 2026 8:50 UTC
5 points
0 comments5 min readLW link
(www.forethought.org)

The Ado­les­cence is Already Here

Priyanka Bharadwaj3 Feb 2026 7:43 UTC
33 points
2 comments2 min readLW link

Ad­dress­ing De­ci­sion The­ory’s Si­mu­la­tion Problem

Ashe Vazquez Nuñez3 Feb 2026 7:02 UTC
11 points
0 comments3 min readLW link

Pa­ter­nal-Nar­ra­tive Ap­proach to AI Alignment

JD Croft3 Feb 2026 3:19 UTC
0 points
1 comment9 min readLW link

Non­prof­its De­serve Bet­ter Operations

Deena Englander3 Feb 2026 2:38 UTC
−2 points
3 comments6 min readLW link

Will AGI ar­rive be­fore the worst cli­mate tip­ping points?

SethW3 Feb 2026 2:36 UTC
13 points
0 comments8 min readLW link
(carboncreatures.substack.com)

In­creas­ing AI Strate­gic Com­pe­tence as a Safety Approach

Wei Dai3 Feb 2026 1:08 UTC
53 points
9 comments1 min readLW link

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

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)