Em­pa­thy Has Out­worn Its Place in Politics

Character#27368 Feb 2026 23:22 UTC
−26 points
8 comments4 min readLW link

The Two-Board Prob­lem: Train­ing En­vi­ron­ment for Re­search Agents

Valerii K.8 Feb 2026 23:13 UTC
4 points
0 comments9 min readLW link

Join My New Move­ment for the Post-AI World

E.G. Blee-Goldman8 Feb 2026 22:18 UTC
0 points
0 comments7 min readLW link

Dona­tions, The Fifth Year

jenn8 Feb 2026 22:04 UTC
39 points
0 comments4 min readLW link
(www.jenn.site)

Every Mea­sure­ment Has a Scale

CarolusRenniusVitellius8 Feb 2026 20:07 UTC
17 points
4 comments4 min readLW link
(charlesr-w.github.io)

UtopiaBench

nielsrolf8 Feb 2026 18:19 UTC
67 points
10 comments1 min readLW link

Smokey, This is not ’Nam Or: [Already] over the [red] line!

Davidmanheim8 Feb 2026 12:24 UTC
110 points
22 comments4 min readLW link

The op­ti­mal age to freeze eggs is 19

GeneSmith8 Feb 2026 9:44 UTC
195 points
48 comments6 min readLW link

It Is Rea­son­able To Re­search How To Use Model In­ter­nals In Training

Neel Nanda8 Feb 2026 3:44 UTC
103 points
15 comments4 min readLW link

Claude’s Bad Primer Fanfic

abramdemski8 Feb 2026 0:39 UTC
24 points
12 comments54 min readLW link

Can thoughtcrimes scare a cau­tious satis­ficer?

Knight Lee7 Feb 2026 23:28 UTC
4 points
4 comments1 min readLW link

[Question] What should I try to do this year?

abstractapplic7 Feb 2026 22:06 UTC
36 points
4 comments1 min readLW link

Does fo­cus­ing on an­i­mal welfare make sense if you’re AI-pilled?

GradientDissenter7 Feb 2026 20:51 UTC
13 points
7 comments8 min readLW link

Near-In­stantly Abort­ing the Worst Pain Imag­in­able with Psychedelics

eleweek7 Feb 2026 16:11 UTC
217 points
13 comments13 min readLW link
(psychotechnology.substack.com)

Why yeast-based vac­cines could be a big deal for biosecurity

delton1377 Feb 2026 16:08 UTC
62 points
8 comments11 min readLW link

Prompt in­jec­tion in Google Trans­late re­veals base model be­hav­iors be­hind task-spe­cific fine-tuning

megasilverfist7 Feb 2026 13:56 UTC
160 points
27 comments3 min readLW link

Eu­nifi­ca­tion: a His­tor­i­cal Perspective

Martin Sustrik7 Feb 2026 13:31 UTC
19 points
5 comments5 min readLW link
(www.250bpm.com)

Vot­ing Re­sults for the 2024 Review

RobertM7 Feb 2026 3:48 UTC
98 points
0 comments1 min readLW link

Play­ing with an In­frared Camera

jefftk7 Feb 2026 3:30 UTC
33 points
1 comment1 min readLW link
(www.jefftk.com)

Honey, I shrunk the brain

Andy_McKenzie7 Feb 2026 0:01 UTC
128 points
1 comment5 min readLW link
(neurobiology.substack.com)

Strat­egy of von Neu­mann and strat­egy of Rosenbergs

avturchin6 Feb 2026 22:50 UTC
5 points
4 comments2 min readLW link

Data-Cen­tric In­ter­pretabil­ity for LLM-based Multi-Agent Re­in­force­ment Learning

6 Feb 2026 19:27 UTC
10 points
0 comments4 min readLW link

Parks Aren’t Nature

Sable6 Feb 2026 18:27 UTC
50 points
11 comments8 min readLW link
(affablyevil.substack.com)

Claude Code #4: From The Be­fore Times

Zvi6 Feb 2026 18:01 UTC
42 points
1 comment23 min readLW link
(thezvi.wordpress.com)

Ro­bust Finite Poli­cies are Non­triv­ially Structured

Winter Cross6 Feb 2026 17:47 UTC
26 points
1 comment11 min readLW link

In (highly con­tin­gent!) defense of in­ter­pretabil­ity-in-the-loop ML training

Steven Byrnes6 Feb 2026 16:32 UTC
85 points
11 comments3 min readLW link

Spec­tral Sig­na­tures of Grad­ual Disempowerment

Jonas Hallgren6 Feb 2026 15:08 UTC
36 points
4 comments17 min readLW link
(equilibria1.substack.com)

De­mands Are All You Need: Prompt Im­per­a­tive­ness Dras­ti­cally Re­duces Hedg­ing In LLMs (n=900, Co­hen’s d = 2.67)

fluxxrider6 Feb 2026 13:22 UTC
6 points
0 comments16 min readLW link

[Question] If all hu­mans were turned into high-fidelity mind up­loads to­mor­row, would we be self-sus­tain­ing?

Erich_Grunewald6 Feb 2026 8:35 UTC
11 points
2 comments1 min readLW link

AI bench­mark­ing has a Y-axis prob­lem

Lizka6 Feb 2026 7:45 UTC
79 points
3 comments7 min readLW link

Claude Opus 4.6 is Driven

HunterJay6 Feb 2026 4:15 UTC
113 points
1 comment5 min readLW link

Why ASI Might Pre­serve Its Progenitors

Luke J. Dawes6 Feb 2026 2:54 UTC
2 points
0 comments12 min readLW link

How Dario Amodei’s “The Ado­les­cence of Tech­nol­ogy” Dele­gi­t­imizes AI X-Risk Concerns

6 Feb 2026 2:07 UTC
12 points
6 comments50 min readLW link
(doomdebates.com)

[Question] Good­fire and Train­ing on Interpretability

Satya Benson6 Feb 2026 1:45 UTC
32 points
5 comments1 min readLW link

Plan ’Straya

William the Kiwi 6 Feb 2026 0:14 UTC
16 points
5 comments5 min readLW link

TT Self Study Jour­nal # 6

TristanTrim5 Feb 2026 23:41 UTC
5 points
3 comments3 min readLW link

The Sim­plest Case for AI Catastrophe

Linch5 Feb 2026 23:18 UTC
77 points
9 comments10 min readLW link
(linch.substack.com)

TT’s Look­ing-for-Work Strategy

TristanTrim5 Feb 2026 21:40 UTC
4 points
0 comments1 min readLW link

Agent Eco­nomics: a BOTEC on feasibility

Margot5 Feb 2026 20:15 UTC
28 points
0 comments6 min readLW link
(forum.effectivealtruism.org)

Molt­book as a set­ting to an­a­lyze Power Seek­ing behaviour

Rahul N5 Feb 2026 20:07 UTC
11 points
0 comments1 min readLW link
(propensitylabs.substack.com)

The na­ture of LLM al­gorith­mic progress (v2)

Steven Byrnes5 Feb 2026 19:17 UTC
116 points
27 comments13 min readLW link

Biotech Startup Stats

sarahconstantin5 Feb 2026 18:40 UTC
21 points
0 comments4 min readLW link
(sarahconstantin.substack.com)

On The Lies De­pres­sion Tells

sonicrocketman5 Feb 2026 17:13 UTC
25 points
2 comments3 min readLW link
(brianschrader.com)

Speedrun­ning a Mech In­terp Re­search Setup (Re­mote GPU, Torch, Trans­formerLens, Cuda, SSH, VS Code)

J Rosser5 Feb 2026 16:45 UTC
38 points
3 comments4 min readLW link

What’s the con­crete plan to be­come an in­cred­ibly agen­tic per­son?

Peter Berggren5 Feb 2026 16:27 UTC
12 points
3 comments3 min readLW link

AI #154: Claw Your Way To The Top

Zvi5 Feb 2026 16:10 UTC
38 points
2 comments43 min readLW link
(thezvi.wordpress.com)

Prepar­ing for a Warn­ing Shot

Noah Birnbaum5 Feb 2026 15:10 UTC
43 points
5 comments4 min readLW link

A Pro­posal for TruesightBench

David Africa5 Feb 2026 14:33 UTC
14 points
0 comments4 min readLW link

Scratch­ing the sore: how plea­sure re­lates to suffering

Vadim Golub5 Feb 2026 12:05 UTC
−1 points
35 comments2 min readLW link

What’s the Point of the Math?

Ashe Vazquez Nuñez5 Feb 2026 11:30 UTC
46 points
3 comments5 min readLW link