Co­op­er­a­tionism: first draft for a moral frame­work that does not re­quire consciousness

Épiphanie Gédéon19 Feb 2026 21:07 UTC
26 points
5 comments8 min readLW link

Flam­in­gos (among other things) re­duce emer­gent misalignment

eekay19 Feb 2026 19:17 UTC
13 points
3 comments7 min readLW link

Funker­ing!

flying buttress19 Feb 2026 18:14 UTC
13 points
0 comments1 min readLW link

Sub­jec­tivity vs Agency: AI “Wak­ing Up”?

Jonathan Moregård19 Feb 2026 17:19 UTC
4 points
0 comments5 min readLW link
(honestliving.substack.com)

You May Already Be Canadian

jefftk19 Feb 2026 16:00 UTC
120 points
14 comments1 min readLW link
(www.jefftk.com)

AI Re­searchers and Ex­ec­u­tives Con­tinue to Un­der­es­ti­mate the Near-Fu­ture Risks of Open Models

Andrew Dickson19 Feb 2026 15:56 UTC
23 points
1 comment16 min readLW link

AI #156 Part 1: They Do Mean The Effect On Jobs

Zvi19 Feb 2026 14:20 UTC
53 points
7 comments36 min readLW link
(thezvi.wordpress.com)

Ter­mi­nal Cynicism

19 Feb 2026 13:51 UTC
24 points
25 comments10 min readLW link
(cognition.cafe)

How much in­for­ma­tion does an op­ti­mal policy con­tain about its en­vi­ron­ment?

19 Feb 2026 13:05 UTC
30 points
0 comments10 min readLW link

All hands on deck to build the dat­a­cen­ter lie detector

Naci Cankaya19 Feb 2026 11:42 UTC
32 points
2 comments5 min readLW link
(open.substack.com)

A Tech­ni­cal Primer on Mechanis­tic Interpretability

Alexei G19 Feb 2026 7:42 UTC
1 point
0 comments11 min readLW link
(alexeigannon.com)

Power Laws Are Not Enough

CarolusRenniusVitellius19 Feb 2026 4:31 UTC
10 points
3 comments4 min readLW link
(charlesr-w.github.io)

Be skep­ti­cal of mile­stone an­nounce­ments by young AI startups

lc19 Feb 2026 4:19 UTC
25 points
0 comments3 min readLW link

Opus 4.5 made a biode­vice (w me)

Raye19 Feb 2026 2:31 UTC
23 points
0 comments10 min readLW link

Re­view of If Any­one Builds It, Every­one Dies

James Brobin19 Feb 2026 1:53 UTC
23 points
4 comments5 min readLW link

I want to ac­tu­ally get good at fore­cast­ing this year (Group In­vite)

Vojtech Brynych19 Feb 2026 1:41 UTC
12 points
4 comments1 min readLW link

Does GPT-2 Rep­re­sent Con­tro­versy? A Small Mech In­terp Investigation

CharlesL19 Feb 2026 1:36 UTC
6 points
0 comments2 min readLW link

Emo­tional Disper­sion and Patience

Astrid Callender19 Feb 2026 1:35 UTC
6 points
5 comments4 min readLW link

What AI-safely top­ics are miss­ing from the main­stream me­dia? What un­der­re­ported but un­der­es­ti­mated is­sues need to be ad­dressed? This is your chance to col­lab­o­rate with film­mak­ers & have your wor­ries ad­dressed.

Max Hellier19 Feb 2026 1:30 UTC
2 points
0 comments1 min readLW link

Man­i­fold spin off MNX, a real money de­cen­tral­ized mar­ket for AI-re­lated bets. In­cludes lev­ered pre­dic­tion mar­kets, per­pet­ual futures

mako yass18 Feb 2026 22:36 UTC
10 points
3 comments1 min readLW link
(x.com)

AI and Na­tion­al­ism Are a Deadly Combination

Matrice Jacobine18 Feb 2026 21:46 UTC
11 points
0 comments4 min readLW link
(www.currentaffairs.org)

Todd, Ord, Galef, Yud­kowsky: Ger­man Pod­cast Sums Up EA/​LW Books

jorges18 Feb 2026 21:44 UTC
7 points
0 comments1 min readLW link

The near-term po­ten­tial of AI fore­cast­ing for pub­lic epistemics

Lawrence Phillips18 Feb 2026 20:37 UTC
21 points
0 comments16 min readLW link

Monthly Roundup #39: Fe­bru­ary 2026

Zvi18 Feb 2026 20:30 UTC
32 points
5 comments40 min readLW link
(thezvi.wordpress.com)

How to Reset

Logan Riggs18 Feb 2026 19:49 UTC
10 points
2 comments2 min readLW link

Karl Pop­per, meet the Hydra

Kotlopou18 Feb 2026 18:55 UTC
14 points
4 comments21 min readLW link
(beatingthehydra.substack.com)

Altru­ism Survey

ozymandias18 Feb 2026 18:40 UTC
9 points
0 comments1 min readLW link

Build­ing Tech­nol­ogy to Drive AI Governance

jsteinhardt18 Feb 2026 18:30 UTC
59 points
4 comments10 min readLW link
(bounded-regret.ghost.io)

Align­ment Is Proven Tractable

SE Gyges18 Feb 2026 17:55 UTC
10 points
0 comments10 min readLW link
(www.verysane.ai)

Why we should ex­pect ruth­less so­ciopath ASI

Steven Byrnes18 Feb 2026 17:28 UTC
163 points
63 comments8 min readLW link

Is the In­visi­ble Hand an Agent?

Gunnar_Zarncke18 Feb 2026 16:26 UTC
13 points
4 comments4 min readLW link
(substack.com)

Nine Fla­vors of Not Enough

Gordon Seidoh Worley18 Feb 2026 15:10 UTC
13 points
0 comments6 min readLW link
(www.uncertainupdates.com)

Grown from Us

ben_levinstein18 Feb 2026 14:57 UTC
10 points
0 comments2 min readLW link

How much su­per­po­si­tion is there?

18 Feb 2026 13:53 UTC
25 points
0 comments3 min readLW link

Ir­ra­tional­ity is So­cially Strategic

Valentine18 Feb 2026 13:28 UTC
119 points
18 comments13 min readLW link

An­nounce­ment: Tech­ni­cal AI Safety Evals Course

18 Feb 2026 13:24 UTC
7 points
0 comments1 min readLW link

Man­aged vs Un­man­aged Agency

plex18 Feb 2026 13:23 UTC
52 points
23 comments3 min readLW link

Ge­nomic eman­ci­pa­tion con­tra eugenics

TsviBT18 Feb 2026 10:35 UTC
56 points
8 comments51 min readLW link

Already Optimized

Florian_Dietz18 Feb 2026 10:01 UTC
52 points
14 comments14 min readLW link

Statis­ti­cal Literacy

kqr18 Feb 2026 6:50 UTC
0 points
2 comments8 min readLW link
(entropicthoughts.com)

AXRP Epi­sode 49 - Cas­par Oester­held on Pro­gram Equilibrium

DanielFilan18 Feb 2026 1:30 UTC
10 points
1 comment72 min readLW link

Thoughts about Understanding

azergante18 Feb 2026 0:19 UTC
4 points
1 comment5 min readLW link

Mon­day AI Radar #13

Against Moloch18 Feb 2026 0:13 UTC
9 points
0 comments8 min readLW link
(againstmoloch.com)

De­cep­tion Chan­nel­ing: Train­ing Models to Always Ver­bal­ize Align­ment Faking

Florian_Dietz17 Feb 2026 22:28 UTC
7 points
2 comments9 min readLW link

Rephras­ing Re­duces Eval Aware­ness...

atharva17 Feb 2026 22:23 UTC
23 points
4 comments3 min readLW link

The Math And The Territory

cylonator17 Feb 2026 21:53 UTC
2 points
0 comments8 min readLW link

Words are not dead

William tirkey17 Feb 2026 21:42 UTC
−2 points
2 comments5 min readLW link

Re­view of the Sys­tem The­ory as a Field of Knowledge

siarshai17 Feb 2026 21:34 UTC
4 points
1 comment18 min readLW link

You’re an AI Ex­pert – Not an Influencer

Max Winga17 Feb 2026 21:03 UTC
180 points
25 comments6 min readLW link
(maxwinga.substack.com)

“We are con­fused about agency”

Cole Wyeth17 Feb 2026 19:51 UTC
57 points
37 comments3 min readLW link