AGI Ruin: A List of Lethalities

Eliezer YudkowskyJun 5, 2022, 10:05 PM
936 points
708 comments30 min readLW link3 reviews

Where I agree and dis­agree with Eliezer

paulfchristianoJun 19, 2022, 7:15 PM
899 points
223 comments18 min readLW link2 reviews

It’s Prob­a­bly Not Lithium

NatáliaJun 28, 2022, 9:24 PM
442 points
187 comments28 min readLW link1 review

Se­cu­rity Mind­set: Les­sons from 20+ years of Soft­ware Se­cu­rity Failures Rele­vant to AGI Alignment

elspoodJun 21, 2022, 11:55 PM
362 points
42 comments7 min readLW link1 review

What Are You Track­ing In Your Head?

johnswentworthJun 28, 2022, 7:30 PM
289 points
83 comments4 min readLW link1 review

A cen­tral AI al­ign­ment prob­lem: ca­pa­bil­ities gen­er­al­iza­tion, and the sharp left turn

So8resJun 15, 2022, 1:10 PM
272 points
55 comments10 min readLW link1 review

Hu­mans are very re­li­able agents

alyssavanceJun 16, 2022, 10:02 PM
269 points
35 comments3 min readLW link

Com­ment re­ply: my low-qual­ity thoughts on why CFAR didn’t get farther with a “real/​effi­ca­cious art of ra­tio­nal­ity”

AnnaSalamonJun 9, 2022, 2:12 AM
261 points
63 comments17 min readLW link1 review

Slow mo­tion videos as AI risk in­tu­ition pumps

Andrew_CritchJun 14, 2022, 7:31 PM
241 points
41 comments2 min readLW link1 review

Con­tra Hofs­tadter on GPT-3 Nonsense

ricticJun 15, 2022, 9:53 PM
237 points
24 comments2 min readLW link

AGI Safety FAQ /​ all-dumb-ques­tions-al­lowed thread

Aryeh EnglanderJun 7, 2022, 5:47 AM
227 points
526 comments4 min readLW link

The pro­to­typ­i­cal catas­trophic AI ac­tion is get­ting root ac­cess to its datacenter

BuckJun 2, 2022, 11:46 PM
180 points
13 comments2 min readLW link1 review

The in­or­di­nately slow spread of good AGI con­ver­sa­tions in ML

Rob BensingerJun 21, 2022, 4:09 PM
173 points
62 comments8 min readLW link

An­nounc­ing the In­verse Scal­ing Prize ($250k Prize Pool)

Jun 27, 2022, 3:58 PM
171 points
14 comments7 min readLW link

AI Could Defeat All Of Us Combined

HoldenKarnofskyJun 9, 2022, 3:50 PM
170 points
42 comments17 min readLW link
(www.cold-takes.com)

On A List of Lethalities

ZviJun 13, 2022, 12:30 PM
165 points
50 comments54 min readLW link1 review
(thezvi.wordpress.com)

A trans­parency and in­ter­pretabil­ity tech tree

evhubJun 16, 2022, 11:44 PM
163 points
11 comments18 min readLW link1 review

Deep Learn­ing Sys­tems Are Not Less In­ter­pretable Than Logic/​Prob­a­bil­ity/​Etc

johnswentworthJun 4, 2022, 5:41 AM
159 points
55 comments2 min readLW link1 review

Godzilla Strategies

johnswentworthJun 11, 2022, 3:44 PM
159 points
72 comments3 min readLW link

Why all the fuss about re­cur­sive self-im­prove­ment?

So8resJun 12, 2022, 8:53 PM
158 points
62 comments7 min readLW link1 review

Limits to Legibility

Jan_KulveitJun 29, 2022, 5:42 PM
157 points
11 comments5 min readLW link1 review

Non­profit Boards are Weird

HoldenKarnofskyJun 23, 2022, 2:40 PM
156 points
26 comments20 min readLW link1 review
(www.cold-takes.com)

LessWrong Has Agree/​Disagree Vot­ing On All New Com­ment Threads

Ben PaceJun 24, 2022, 12:43 AM
154 points
217 comments2 min readLW link1 review

Stay­ing Split: Sa­ba­tini and So­cial Justice

Duncan Sabien (Deactivated)Jun 8, 2022, 8:32 AM
153 points
28 comments21 min readLW link

Steam

abramdemskiJun 20, 2022, 5:38 PM
149 points
13 comments5 min readLW link1 review

[Question] why as­sume AGIs will op­ti­mize for fixed goals?

nostalgebraistJun 10, 2022, 1:28 AM
147 points
60 comments4 min readLW link2 reviews

Public be­liefs vs. Pri­vate beliefs

Eli TyreJun 1, 2022, 9:33 PM
144 points
30 comments5 min readLW link

A de­scrip­tive, not pre­scrip­tive, overview of cur­rent AI Align­ment Research

Jun 6, 2022, 9:59 PM
139 points
21 comments7 min readLW link

An­nounc­ing the LessWrong Cu­rated Podcast

Jun 22, 2022, 10:16 PM
137 points
27 comments1 min readLW link

AI-Writ­ten Cri­tiques Help Hu­mans No­tice Flaws

paulfchristianoJun 25, 2022, 5:22 PM
137 points
5 comments3 min readLW link
(openai.com)

Con­tra EY: Can AGI de­stroy us with­out trial & er­ror?

nsokolskyJun 13, 2022, 6:26 PM
137 points
72 comments15 min readLW link

Will Ca­pa­bil­ities Gen­er­al­ise More?

Ramana KumarJun 29, 2022, 5:12 PM
133 points
39 comments4 min readLW link

In­ter­gen­er­a­tional trauma im­ped­ing co­op­er­a­tive ex­is­ten­tial safety efforts

Andrew_CritchJun 3, 2022, 8:13 AM
129 points
29 comments3 min readLW link

Con­fused why a “ca­pa­bil­ities re­search is good for al­ign­ment progress” po­si­tion isn’t dis­cussed more

Kaj_SotalaJun 2, 2022, 9:41 PM
129 points
27 comments4 min readLW link

“Pivotal Acts” means some­thing specific

RaemonJun 7, 2022, 9:56 PM
127 points
23 comments2 min readLW link

Let’s See You Write That Cor­rigi­bil­ity Tag

Eliezer YudkowskyJun 19, 2022, 9:11 PM
124 points
70 comments1 min readLW link

Scott Aaron­son is join­ing OpenAI to work on AI safety

peterbarnettJun 18, 2022, 4:06 AM
117 points
31 comments1 min readLW link
(scottaaronson.blog)

CFAR Hand­book: Introduction

CFAR!DuncanJun 28, 2022, 4:53 PM
116 points
12 comments1 min readLW link

Leav­ing Google, Join­ing the Nu­cleic Acid Observatory

jefftkJun 10, 2022, 5:00 PM
114 points
4 comments3 min readLW link
(www.jefftk.com)

Con­ver­sa­tion with Eliezer: What do you want the sys­tem to do?

Orpheus16Jun 25, 2022, 5:36 PM
114 points
38 comments2 min readLW link

Who mod­els the mod­els that model mod­els? An ex­plo­ra­tion of GPT-3′s in-con­text model fit­ting ability

LovreJun 7, 2022, 7:37 PM
112 points
16 comments9 min readLW link

Re­la­tion­ship Ad­vice Repository

RubyJun 20, 2022, 2:39 PM
109 points
36 comments38 min readLW link

wrap­per-minds are the enemy

nostalgebraistJun 17, 2022, 1:58 AM
104 points
43 comments8 min readLW link

Yes, AI re­search will be sub­stan­tially cur­tailed if a lab causes a ma­jor disaster

lcJun 14, 2022, 10:17 PM
103 points
31 comments2 min readLW link

The Moun­tain Troll

lsusrJun 11, 2022, 9:14 AM
103 points
26 comments2 min readLW link

Units of Exchange

CFAR!DuncanJun 28, 2022, 4:53 PM
99 points
28 comments11 min readLW link

Pivotal out­comes and pivotal processes

Andrew_CritchJun 17, 2022, 11:43 PM
97 points
31 comments4 min readLW link

An­nounc­ing Epoch: A re­search or­ga­ni­za­tion in­ves­ti­gat­ing the road to Trans­for­ma­tive AI

Jun 27, 2022, 1:55 PM
97 points
2 comments2 min readLW link
(epochai.org)

My cur­rent take on In­ter­nal Fam­ily Sys­tems “parts”

Kaj_SotalaJun 26, 2022, 5:40 PM
96 points
11 comments3 min readLW link
(kajsotala.fi)

Con­test: An Alien Message

DaemonicSigilJun 27, 2022, 5:54 AM
95 points
100 comments1 min readLW link