Let’s See You Write That Cor­rigi­bil­ity Tag

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

Half-baked al­ign­ment idea: train­ing to generalize

Aaron BergmanJun 19, 2022, 8:16 PM
10 points
2 comments4 min readLW link

Where I agree and dis­agree with Eliezer

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

[Question] AI mis­al­ign­ment risk from GPT-like sys­tems?

fiso64Jun 19, 2022, 5:35 PM
10 points
8 comments1 min readLW link

[Link-post] On Defer­ence and Yud­kowsky’s AI Risk Estimates

bmgJun 19, 2022, 5:25 PM
29 points
8 comments1 min readLW link

Heb­bian Learn­ing Is More Com­mon Than You Think

Aleksi LiimatainenJun 19, 2022, 3:57 PM
8 points
2 comments1 min readLW link

The Malthu­sian Trap: An Ex­tremely Short Introduction

Davis KedroskyJun 19, 2022, 3:25 PM
5 points
0 comments6 min readLW link
(daviskedrosky.substack.com)

Par­li­a­ments with­out the Parties

Yair HalberstadtJun 19, 2022, 2:06 PM
18 points
18 comments2 min readLW link

Lamda is not an LLM

KevinJun 19, 2022, 11:13 AM
7 points
10 comments1 min readLW link
(www.wired.com)

Get­ting stuck in lo­cal minima

louis030195Jun 19, 2022, 8:50 AM
3 points
1 comment1 min readLW link
(brain.louis030195.com)

[Linkpost] The im­por­tance of stu­pidity in sci­en­tific research

PatternJun 19, 2022, 5:17 AM
17 points
1 comment1 min readLW link
(journals.biologists.com)

ETH is prob­a­bly un­der­val­ued right now

mukashiJun 19, 2022, 2:20 AM
−7 points
22 comments1 min readLW link

Juneberry Cake

jefftkJun 19, 2022, 1:40 AM
29 points
0 comments1 min readLW link
(www.jefftk.com)

Agent level parallelism

Johannes C. MayerJun 18, 2022, 8:56 PM
5 points
5 comments1 min readLW link

What are our outs to play to?

HastingsJun 18, 2022, 7:32 PM
7 points
0 comments2 min readLW link

[Question] What’s the in­for­ma­tion value of gov­ern­ment hear­ings?

KennyJun 18, 2022, 5:13 PM
6 points
4 comments2 min readLW link

The best ‘free solo’ (rock climb­ing) video

KennyJun 18, 2022, 3:29 PM
14 points
4 comments2 min readLW link

[Question] What’s the name of this fal­lacy/​rea­son­ing an­tipat­tern?

David GrossJun 18, 2022, 2:04 PM
9 points
6 comments1 min readLW link

“Brain en­thu­si­asts” in AI Safety

Jun 18, 2022, 9:59 AM
63 points
5 comments10 min readLW link
(universalprior.substack.com)

To what ex­tent have ideas and sci­en­tific dis­cov­er­ies got­ten harder to find?

lsusrJun 18, 2022, 7:15 AM
33 points
10 comments6 min readLW link

[Question] What’s the goal in life?

Konstantin WeitzJun 18, 2022, 6:09 AM
5 points
6 comments1 min readLW link

Can DALL-E un­der­stand sim­ple ge­om­e­try?

Isaac KingJun 18, 2022, 4:37 AM
25 points
2 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)

[Question] Why don’t we think we’re in the sim­plest uni­verse with in­tel­li­gent life?

ADifferentAnonymousJun 18, 2022, 3:05 AM
30 points
33 comments1 min readLW link

Do your­self a FAVAR: se­cu­rity mindset

lemonhopeJun 18, 2022, 2:08 AM
20 points
2 comments2 min readLW link

Fore­cast­ing Fu­sion Power

Daniel KokotajloJun 18, 2022, 12:04 AM
29 points
8 comments1 min readLW link
(astralcodexten.substack.com)

Pivotal out­comes and pivotal processes

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

Quan­tify­ing Gen­eral Intelligence

JasonBrownJun 17, 2022, 9:57 PM
9 points
6 comments13 min readLW link

Ap­ply for Pro­duc­tivity Coach­ing and AI Align­ment Mentorship

NickJun 17, 2022, 9:36 PM
12 points
1 comment1 min readLW link

Things That Make Me En­joy Giv­ing Ca­reer Advice

Neel NandaJun 17, 2022, 8:49 PM
16 points
0 comments9 min readLW link
(www.neelnanda.io)

The Unified The­ory of Nor­ma­tive Ethics

Thane RuthenisJun 17, 2022, 7:55 PM
8 points
0 comments6 min readLW link

1689: Un­cov­er­ing the World New In­sti­tu­tion­al­ism Created

Davis KedroskyJun 17, 2022, 7:32 PM
7 points
0 comments9 min readLW link
(daviskedrosky.substack.com)

[Question] Is there an unified way to make sense of ai failure modes?

walking_mushroomJun 17, 2022, 6:00 PM
3 points
1 comment1 min readLW link

In defense of flailing, with fore­word by Bill Burr

lcJun 17, 2022, 4:40 PM
88 points
6 comments4 min readLW link

An Ap­proach to Land Value Taxation

harsimonyJun 17, 2022, 3:53 PM
4 points
12 comments4 min readLW link
(harsimony.wordpress.com)

Value ex­trap­o­la­tion vs Wireheading

Stuart_ArmstrongJun 17, 2022, 3:02 PM
16 points
1 comment1 min readLW link

#SAT with Ten­sor Networks

Adam JermynJun 17, 2022, 1:20 PM
4 points
0 comments2 min readLW link

An­nounc­ing the Clearer Think­ing Re­grants program

spencergJun 17, 2022, 1:14 PM
36 points
1 comment1 min readLW link

Sin­ga­pore—Small ca­sual din­ner in Chi­na­town #3: DALL-E 2 edition

Joe RoccaJun 17, 2022, 8:32 AM
2 points
2 comments1 min readLW link

[Question] Is civ­i­liza­tional al­ign­ment on the table?

Aleksi LiimatainenJun 17, 2022, 8:27 AM
5 points
1 comment1 min readLW link

Ap­ply to the Ma­chine Learn­ing For Good boot­camp in France

Alexandre VariengienJun 17, 2022, 7:32 AM
10 points
0 comments1 min readLW link

What’s it like to have sex with Dun­can?

Duncan Sabien (Inactive)Jun 17, 2022, 2:32 AM
52 points
19 comments17 min readLW link

wrap­per-minds are the enemy

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

A Li­tany Miss­ing from the Canon

benwrJun 17, 2022, 1:39 AM
39 points
3 comments1 min readLW link
(www.benwr.net)

[Question] Why did Rus­sia in­vade Ukraine?

bohaskaJun 17, 2022, 1:36 AM
0 points
5 comments1 min readLW link

A trans­parency and in­ter­pretabil­ity tech tree

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

BBC Fu­ture cov­ers progress studies

jasoncrawfordJun 16, 2022, 10:44 PM
21 points
6 comments3 min readLW link
(rootsofprogress.org)

Hu­mans are very re­li­able agents

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

Towards Gears-Level Un­der­stand­ing of Agency

Thane RuthenisJun 16, 2022, 10:00 PM
25 points
4 comments18 min readLW link

A pos­si­ble AI-in­oc­u­la­tion due to early “robot up­ris­ing”

ShmiJun 16, 2022, 9:21 PM
16 points
2 comments1 min readLW link