✕
RSS

Qualia-based emo­tion steer­ing makes LLMs at­tribute con­scious states to themselves

agastyasridharan6 Jul 2026 10:24 UTC
11 points
0 comments10 min readLW link

Com­pu­ta­tion en­ables Ac­tion: Ex­plod­ing the Si­mu­la­tion Fallacy

ChrisHibbert6 Jul 2026 2:54 UTC
24 points
0 comments6 min readLW link

Coun­ter­fac­tual mug­ging is a limit­ing case of Psy-kosh’s non-an­thropic problem

Abhimanyu Pallavi Sudhir6 Jul 2026 2:21 UTC
5 points
0 comments2 min readLW link

Reflec­tions on The Scout Mindset

James Brobin5 Jul 2026 20:18 UTC
10 points
0 comments3 min readLW link

When Gemma Thinks About Re­sources—it Fails: a Be­hav­ioral Experiment

TheVinci5 Jul 2026 19:23 UTC
8 points
1 comment2 min readLW link
(tarantulabs.com)

Claude’s mal­i­cious com­pli­ance and nor­mal­iza­tion of deviance

Steff5 Jul 2026 17:18 UTC
34 points
18 comments8 min readLW link

Prob­ing the loss-band spar­sity as­sump­tion in Scien­tist AI

Alejandro Tlaie5 Jul 2026 16:25 UTC
16 points
0 comments7 min readLW link

Book Re­view: The God Test

PeterMcCluskey5 Jul 2026 16:11 UTC
21 points
1 comment4 min readLW link

We need 3rd party Train­ing-Run Assessments

Alex Meinke5 Jul 2026 15:55 UTC
93 points
0 comments10 min readLW link

Harry Pot­ter and the Rules of Quidditch

Tomás B.5 Jul 2026 14:32 UTC
129 points
3 comments3 min readLW link

A Nor­mal Ar­gu­ment for AI Risk

Silent Swift5 Jul 2026 9:32 UTC
14 points
3 comments8 min readLW link
(silentswift.substack.com)

The case for the fleshman

dr_s5 Jul 2026 9:28 UTC
16 points
17 comments4 min readLW link

Ree­val­u­at­ing AI-2027: timelines, take­off, al­ign­ment and China

StanislavKrym5 Jul 2026 4:00 UTC
14 points
6 comments5 min readLW link

Suc­cess Per Tokens

michaelwaves5 Jul 2026 2:25 UTC
8 points
0 comments3 min readLW link

A case for LLMs as Self-predictors

Ashe Vazquez Nuñez5 Jul 2026 0:25 UTC
31 points
1 comment10 min readLW link

Defin­ing in­ter­pre­ta­tion, and es­tab­lish­ing a frame­work for it

Yaroven4 Jul 2026 16:31 UTC
6 points
0 comments5 min readLW link

The Lace (short story)

Michael Soareverix4 Jul 2026 4:43 UTC
24 points
0 comments4 min readLW link

Ap­prox­i­mate Nat­u­ral La­tents Have Ex­act Prices

Haru4 Jul 2026 1:57 UTC
19 points
0 comments6 min readLW link

I think al­ign­ment work is more promis­ing than con­trol work

Alec Harris3 Jul 2026 23:40 UTC
88 points
15 comments8 min readLW link

On “gen­dertropes” in dath ilan

Eliezer Yudkowsky3 Jul 2026 22:20 UTC
58 points
1 comment3 min readLW link