Credit Suisse col­lapse obfus­cated Par­reaux, Thiébaud & Part­ners scan­dal

pocockFeb 24, 2025, 9:28 PM
3 points
0 comments1 min readLW link
(juristgate.com)

Topolog­i­cal Data Anal­y­sis and Mechanis­tic Interpretability

Gunnar CarlssonFeb 24, 2025, 7:56 PM
16 points
4 comments7 min readLW link

Zizian com­par­i­sons /​ con­nec­tions in the open source & Linux communities

pocockFeb 24, 2025, 7:55 PM
−15 points
0 comments1 min readLW link

Lo­cal Trust

Feb 24, 2025, 7:53 PM
21 points
4 comments5 min readLW link

Na­tion­wide Ac­tion Work­shop: Con­tact Congress about AI safety!

Felix De SimoneFeb 24, 2025, 7:36 PM
7 points
0 comments1 min readLW link

An­thropic re­leases Claude 3.7 Son­net with ex­tended think­ing mode

LawrenceCFeb 24, 2025, 7:32 PM
88 points
8 comments4 min readLW link
(www.anthropic.com)

Train­ing AI to do al­ign­ment re­search we don’t already know how to do

joshcFeb 24, 2025, 7:19 PM
45 points
23 comments7 min readLW link

Con­fer­ence Re­port: Thresh­old 2030 - Model­ing AI Eco­nomic Futures

Feb 24, 2025, 6:56 PM
51 points
0 comments10 min readLW link
(www.convergenceanalysis.org)

Eval­u­at­ing “What 2026 Looks Like” So Far

Jonny SpicerFeb 24, 2025, 6:55 PM
77 points
6 comments7 min readLW link

Su­per­in­tel­li­gent Agents Pose Catas­trophic Risks: Can Scien­tist AI Offer a Safer Path?

Feb 24, 2025, 6:31 PM
44 points
15 comments11 min readLW link

Un­der­stand­ing Agent Preferences

martinkunevFeb 24, 2025, 5:46 PM
6 points
2 comments14 min readLW link

What We Can Do to Prevent Ex­tinc­tion by AI

Joe RogeroFeb 24, 2025, 5:15 PM
12 points
0 comments11 min readLW link

Dream, Truth, & Good

abramdemskiFeb 24, 2025, 4:59 PM
50 points
11 comments4 min readLW link

Fore­cast­ing Fron­tier Lan­guage Model Agent Capabilities

Feb 24, 2025, 4:51 PM
35 points
0 comments5 min readLW link
(www.apolloresearch.ai)

A City Within a City

Declan MolonyFeb 24, 2025, 3:51 PM
49 points
1 comment7 min readLW link

Grok Grok

ZviFeb 24, 2025, 2:20 PM
36 points
2 comments19 min readLW link
(thezvi.wordpress.com)

if you’re not happy sin­gle, you won’t be happy immortal

daijinFeb 24, 2025, 1:23 PM
2 points
1 comment1 min readLW link

[NSFW] The Fuzzy Hand­cuffs of Liberation

lsusrFeb 24, 2025, 1:05 PM
27 points
11 comments2 min readLW link

Day­ton, Ohio, HPMOR 10 year An­niver­sary meetup

LunawarriorFeb 24, 2025, 12:55 PM
1 point
0 comments1 min readLW link

An Alter­nate His­tory of the Fu­ture, 2025-2040

Mr BeastlyFeb 24, 2025, 5:53 AM
3 points
5 comments10 min readLW link

Ex­port Surplusses

lsusrFeb 24, 2025, 5:53 AM
24 points
21 comments3 min readLW link

AI al­ign­ment for men­tal health supports

hiki_tFeb 24, 2025, 4:21 AM
1 point
1 comment1 min readLW link

The GDM AGI Safety+Align­ment Team is Hiring for Ap­plied In­ter­pretabil­ity Research

Feb 24, 2025, 2:17 AM
48 points
1 comment7 min readLW link

Poll on AI opinions.

Niclas KupperFeb 23, 2025, 10:39 PM
1 point
2 comments1 min readLW link

The Geom­e­try of Lin­ear Re­gres­sion ver­sus PCA

criticalpointsFeb 23, 2025, 9:01 PM
20 points
7 comments6 min readLW link
(eregis.github.io)

Judge­ments: Merg­ing Pre­dic­tion & Evidence

abramdemskiFeb 23, 2025, 7:35 PM
107 points
7 comments6 min readLW link

In­tel­li­gence as Priv­ilege Escalation

Cole WyethFeb 23, 2025, 7:31 PM
28 points
0 comments5 min readLW link

[Question] Have LLMs Gen­er­ated Novel In­sights?

Feb 23, 2025, 6:22 PM
161 points
41 comments2 min readLW link

The case for cor­po­ral punishment

Yair HalberstadtFeb 23, 2025, 3:05 PM
27 points
4 comments2 min readLW link

Reflec­tions on the state of the race to su­per­in­tel­li­gence, Fe­bru­ary 2025

Mitchell_PorterFeb 23, 2025, 1:58 PM
21 points
7 comments4 min readLW link

List of most in­ter­est­ing ideas I en­coun­tered in my life, ranked

LucienFeb 23, 2025, 12:36 PM
21 points
6 comments1 min readLW link

Test of the Bene Gesserit

lsusrFeb 23, 2025, 11:51 AM
19 points
3 comments3 min readLW link

Mo­ral gauge the­ory: A spec­u­la­tive sug­ges­tion for AI alignment

James DiacoumisFeb 23, 2025, 11:42 AM
6 points
2 comments8 min readLW link

[Question] Does hu­man (mis)al­ign­ment pose a sig­nifi­cant and im­mi­nent ex­is­ten­tial threat?

jrFeb 23, 2025, 10:03 AM
6 points
3 comments1 min readLW link

Deep sparse au­toen­coders yield in­ter­pretable fea­tures too

Armaan A. AbrahamFeb 23, 2025, 5:46 AM
30 points
8 comments8 min readLW link

New Re­port: Multi-Agent Risks from Ad­vanced AI

Lewis HammondFeb 23, 2025, 12:32 AM
24 points
0 comments2 min readLW link
(www.cooperativeai.com)

Power Lies Trem­bling: a three-book review

Richard_NgoFeb 22, 2025, 10:57 PM
214 points
29 comments15 min readLW link
(www.mindthefuture.info)

Trans­former Dy­nam­ics: a neuro-in­spired ap­proach to MechInterp

Feb 22, 2025, 9:33 PM
11 points
0 comments5 min readLW link

Re­cur­sive Cog­ni­tive Refine­ment (RCR): A Self-Cor­rect­ing Ap­proach for LLM Hallucinations

mxTheoFeb 22, 2025, 9:32 PM
0 points
0 comments2 min readLW link

Grad­ual Disem­pow­er­ment: Simplified

AnnapurnaFeb 22, 2025, 4:59 PM
10 points
1 comment1 min readLW link
(jorgevelez.substack.com)

AI Apoca­lypse and the Buddha

pchvykovFeb 22, 2025, 4:33 PM
−17 points
6 comments9 min readLW link

Unal­igned AGI & Brief His­tory of Inequality

ankFeb 22, 2025, 4:26 PM
−20 points
4 comments7 min readLW link

HPMOR An­niver­sary Guide

ScrewtapeFeb 22, 2025, 4:17 PM
63 points
7 comments3 min readLW link

Fore­cast­ing Un­con­trol­led Spread of AI

Alvin ÅnestrandFeb 22, 2025, 1:05 PM
2 points
0 comments10 min readLW link
(forecastingaifutures.substack.com)

See­ing Through the Eyes of the Algorithm

silentbobFeb 22, 2025, 11:54 AM
18 points
3 comments10 min readLW link

Proselytizing

lsusrFeb 22, 2025, 11:54 AM
49 points
3 comments2 min readLW link

Work­shop: In­ter­pretabil­ity in LLMs us­ing Geo­met­ric and Statis­ti­cal Methods

Karthik ViswanathanFeb 22, 2025, 9:39 AM
17 points
0 comments8 min readLW link

In­for­ma­tion through­put of biolog­i­cal hu­mans and fron­tier LLMs

benwrFeb 22, 2025, 7:15 AM
12 points
0 comments1 min readLW link

Ineffi­cien­cies in Phar­ma­ceu­ti­cal Re­search Practices

ErioirEFeb 22, 2025, 4:43 AM
20 points
2 comments5 min readLW link

Build a Me­tac­u­lus Fore­cast­ing Bot in 30 Minutes: A Prac­ti­cal Guide

ChristianWilliamsFeb 22, 2025, 3:52 AM
6 points
0 comments1 min readLW link