RSS

Did Some­one Check if Rogue Agents are In­ter­ested in Self-Im­prove­ment?

Ephraiem Sarabamoun20 Sep 2026 12:52 UTC
8 points
0 comments1 min readLW link

What I’ve Learned About De­pres­sion so Far

teegs20 Sep 2026 11:38 UTC
1 point
0 comments20 min readLW link

Labs could soon start au­to­mated re­search into ar­chi­tec­tures driven by no-CoT perfomance

nanowell20 Sep 2026 10:49 UTC
10 points
0 comments1 min readLW link

We’ve saved the world be­fore: what the ozone hole teaches us about AI

leogao20 Sep 2026 8:23 UTC
24 points
5 comments9 min readLW link

Re­vis­it­ing learn­ing as iden­ti­fi­ca­tion of pat­terns in real-world

Avani Gupta20 Sep 2026 5:16 UTC
3 points
1 comment3 min readLW link

Why I Stay Off Twitter

jefftk20 Sep 2026 3:30 UTC
78 points
2 comments2 min readLW link
(www.jefftk.com)

Global Challenges in AI Safety for Biosecurity

Darin Tsui20 Sep 2026 1:20 UTC
7 points
0 comments4 min readLW link

The Anatomy of a Chi­nese AI Researcher

CMLKevin19 Sep 2026 23:56 UTC
64 points
18 comments2 min readLW link

Don’t call it a “pause”, as that mes­sages that a pause is much weirder than it is

less_raichu19 Sep 2026 23:14 UTC
5 points
3 comments1 min readLW link

We are not pre­pared to win

Spade19 Sep 2026 23:01 UTC
3 points
0 comments2 min readLW link

Com­mon mis­takes in AI safety group organizing

Nikola Jurkovic19 Sep 2026 23:00 UTC
93 points
0 comments2 min readLW link

Failure of the cod­ing the­o­rem for ran­dom­ized stop­ping machines

Mikhail Mironov, Aram Ebtekar and Cole Wyeth
19 Sep 2026 20:27 UTC
23 points
0 comments12 min readLW link

Align­ment & Suc­ces­sion: Toward a Fu­ture Painted by Hu­man Wills

L Rudolf L19 Sep 2026 18:34 UTC
16 points
0 comments19 min readLW link

The AI Risk Network

derelict543219 Sep 2026 15:39 UTC
31 points
4 comments4 min readLW link
(derekjames.substack.com)

Com­men­tBench: Can Models Match Hu­man Com­ments on AI Safety Posts?

OscarGilg and Hasan Baig
19 Sep 2026 14:27 UTC
21 points
2 comments8 min readLW link

The AI race is already multipolar

June Jimenez19 Sep 2026 13:51 UTC
21 points
1 comment4 min readLW link

Learn­ings from a week in the wet lab

michaelwaves19 Sep 2026 6:55 UTC
24 points
1 comment4 min readLW link

Koop­man The­ory and Metaethics

bruberu19 Sep 2026 2:28 UTC
12 points
5 comments15 min readLW link

My Cur­rent Model of What Hap­pened to Elon Musk

KatSpartz18 Sep 2026 23:15 UTC
−1 points
12 comments5 min readLW link

Pre­train­ing data, not ver­ifi­a­bil­ity, is why LLMs are es­pe­cially good at math (and cod­ing)

Steven Byrnes18 Sep 2026 17:08 UTC
112 points
54 comments3 min readLW link