How an AI com­pany CEO could quietly take over the world

Alex Kastner23 Oct 2025 23:33 UTC
52 points
13 comments11 min readLW link

Wor­lds Where Iter­a­tive De­sign Suc­ceeds?

Max Harms23 Oct 2025 22:14 UTC
23 points
5 comments8 min readLW link

Au­to­mated real time mon­i­tor­ing and or­ches­tra­tion of cod­ing agents

23 Oct 2025 22:12 UTC
8 points
0 comments2 min readLW link
(fulcrumresearch.ai)

Re­minder: Mo­ral­ity is unsolved

Jesper L.23 Oct 2025 21:42 UTC
27 points
45 comments3 min readLW link

The main way I’ve seen peo­ple turn ide­olog­i­cally crazy [Linkpost]

Noosphere8923 Oct 2025 20:09 UTC
123 points
22 comments8 min readLW link
(andymasley.substack.com)

Em­piri­cal Par­tial Derivatives

sonicrocketman23 Oct 2025 17:54 UTC
8 points
0 comments3 min readLW link
(brianschrader.com)

Build­ing a differ­ent kind of per­sonal intelligence

Rebecca Dai23 Oct 2025 17:45 UTC
7 points
0 comments9 min readLW link
(rebeccadai.substack.com)

Beliefs about for­mal meth­ods and AI safety

Quinn23 Oct 2025 16:43 UTC
32 points
0 comments5 min readLW link

AI #139: The Over­reach Machines

Zvi23 Oct 2025 15:30 UTC
35 points
5 comments52 min readLW link
(thezvi.wordpress.com)

Should AI Devel­op­ers Re­move Dis­cus­sion of AI Misal­ign­ment from AI Train­ing Data?

Alek Westover23 Oct 2025 15:12 UTC
43 points
3 comments9 min readLW link

Se­cureBio is Hiring Soft­ware Engineers

jefftk23 Oct 2025 14:10 UTC
21 points
0 comments1 min readLW link
(www.jefftk.com)

Is ter­mi­nal lu­cidity real?

Ariel Zeleznikow-Johnston23 Oct 2025 11:40 UTC
20 points
0 comments1 min readLW link
(open.substack.com)

A Con­crete Roadmap to­wards Safety Cases based on Chain-of-Thought Monitoring

Wuschel Schulz23 Oct 2025 11:34 UTC
37 points
5 comments4 min readLW link
(arxiv.org)

Differ­ences in Align­ment Be­havi­our be­tween Sin­gle-Agent and Multi-Agent AI Systems

23 Oct 2025 11:17 UTC
7 points
3 comments5 min readLW link

LW Psychosis

Annabelle23 Oct 2025 8:12 UTC
18 points
10 comments3 min readLW link

An­nounc­ing the Fu­turekind Win­ter Fel­low­ship 2025/​6

Aditya S23 Oct 2025 5:40 UTC
1 point
0 comments4 min readLW link

Learn­ing to In­ter­pret Weight Differ­ences in Lan­guage Models

avichal23 Oct 2025 3:55 UTC
89 points
2 comments5 min readLW link
(arxiv.org)

AGI’s Last Bottlenecks

adamk23 Oct 2025 3:28 UTC
17 points
2 comments9 min readLW link

State­ment on Su­per­in­tel­li­gence—FLI Open Letter

plex22 Oct 2025 22:26 UTC
59 points
0 comments1 min readLW link
(superintelligence-statement.org)

The Doomers Were Right

Algon22 Oct 2025 22:18 UTC
204 points
26 comments3 min readLW link

Tech­ni­cal Ac­cel­er­a­tion Meth­ods for AI Safety: Sum­mary from Oc­to­ber 2025 Symposium

Martin Leitgab22 Oct 2025 21:33 UTC
25 points
2 comments6 min readLW link

Why AI al­ign­ment mat­ters today

Mislav Jurić22 Oct 2025 21:27 UTC
6 points
0 comments4 min readLW link

Any cor­rigi­bil­ity naysay­ers out­side of MIRI?

Max Harms22 Oct 2025 21:26 UTC
28 points
24 comments1 min readLW link

Which side of the AI safety com­mu­nity are you in?

Max Tegmark22 Oct 2025 21:17 UTC
141 points
88 comments2 min readLW link

Ho­mo­mor­phi­cally en­crypted con­scious­ness and its implications

jessicata22 Oct 2025 20:27 UTC
35 points
48 comments12 min readLW link
(unstableontology.com)

Dead-switches as AI safety tools

Jesper L.22 Oct 2025 19:57 UTC
2 points
6 comments5 min readLW link

Con­sider donat­ing to AI safety cham­pion Scott Wiener

Eric Neyman22 Oct 2025 18:40 UTC
133 points
9 comments18 min readLW link
(ericneyman.wordpress.com)

Pos­tra­tional­ity: An Oral History

Gordon Seidoh Worley22 Oct 2025 16:10 UTC
44 points
4 comments30 min readLW link
(www.uncertainupdates.com)

Penny’s Hands

Tomás B.22 Oct 2025 16:09 UTC
70 points
7 comments16 min readLW link

Is 90% of code at An­thropic be­ing writ­ten by AIs?

ryan_greenblatt22 Oct 2025 14:50 UTC
91 points
14 comments5 min readLW link

How Well Does RL Scale?

Toby_Ord22 Oct 2025 13:16 UTC
131 points
22 comments7 min readLW link

LLM Self-Refer­ence Lan­guage in Mul­tilin­gual vs English-Cen­tric Models

dwmd22 Oct 2025 12:44 UTC
4 points
0 comments6 min readLW link

The Cloud in­dus­try ar­chi­tec­ture [In­fra-Plat­form-App] is un­likely to repli­cate for AI

Armchair Descending22 Oct 2025 8:28 UTC
1 point
0 comments2 min readLW link

The Per­pet­ual Tech­nolog­i­cal Cage

Hector Perez Arenas22 Oct 2025 8:15 UTC
6 points
2 comments1 min readLW link
(networksocieties.com)

Uto­pi­og­ra­phy Interview

plex22 Oct 2025 8:03 UTC
32 points
0 comments45 min readLW link

White House OSTP AI Dereg­u­la­tion Public Com­ment Pe­riod Ends Oct. 27

Zack_M_Davis22 Oct 2025 6:18 UTC
42 points
1 comment1 min readLW link

July-Oc­to­ber 2025 Progress in Guaran­teed Safe AI

Quinn22 Oct 2025 2:30 UTC
15 points
2 comments7 min readLW link
(gsai.substack.com)

In re­mem­brance of Son­net ‘3.6’

kromem22 Oct 2025 0:43 UTC
14 points
9 comments2 min readLW link

Strat­ified Utopia

Cleo Nardo21 Oct 2025 19:09 UTC
73 points
8 comments11 min readLW link

Early stage goal-directednesss

Raemon21 Oct 2025 17:41 UTC
20 points
8 comments3 min readLW link

On Dwarkesh Pa­tel’s Pod­cast With An­drej Karpathy

Zvi21 Oct 2025 16:00 UTC
38 points
6 comments31 min readLW link
(thezvi.wordpress.com)

Sa­muel x Bhishma—Su­per­in­tel­li­gence by 2030?

samuelshadrach21 Oct 2025 15:32 UTC
6 points
0 comments3 min readLW link
(youtu.be)

Re­marks on Bayesian stud­ies from 1963

dynomight21 Oct 2025 12:47 UTC
37 points
1 comment1 min readLW link

Why deep space pro­grams se­lect for calm agree­able in­tro­verted candidates

David Sun21 Oct 2025 10:22 UTC
−4 points
0 comments15 min readLW link

⿻ Sym­bio­ge­n­e­sis vs. Con­ver­gent Consequentialism

21 Oct 2025 10:10 UTC
60 points
5 comments20 min readLW link

How the Hu­man Lens Shapes Ma­chine Minds

21 Oct 2025 9:08 UTC
2 points
0 comments5 min readLW link

21st Cen­tury Civ­i­liza­tion curriculum

Richard_Ngo21 Oct 2025 7:43 UTC
35 points
10 comments1 min readLW link
(www.21civ.com)

Ram­blings on the Self Indi­ca­tion Assumption

Angela Pretorius21 Oct 2025 5:45 UTC
5 points
1 comment2 min readLW link

An epistemic the­ory of pop­ulism [link post to Joseph Heath]

Siebe21 Oct 2025 5:30 UTC
12 points
3 comments1 min readLW link
(open.substack.com)

EU ex­plained in 10 minutes

Martin Sustrik21 Oct 2025 4:40 UTC
244 points
49 comments8 min readLW link
(www.250bpm.com)