Con­di­tion­ing Gen­er­a­tive Models

Adam JermynJun 25, 2022, 10:15 PM
24 points
18 comments10 min readLW link

Unforgivable

NovalisJun 25, 2022, 8:57 PM
−9 points
12 comments5 min readLW link
(novalis.blog)

SunPJ in Alenia

FlorianHJun 25, 2022, 7:39 PM
9 points
19 comments8 min readLW link
(plausiblestuff.com)

[Question] Should any hu­man en­slave an AGI sys­tem?

AlignmentMirrorJun 25, 2022, 7:35 PM
−13 points
44 comments1 min readLW link

Fun­da­men­tal Uncer­tainty: Chap­ter 3 - Why don’t we agree on what’s right?

Gordon Seidoh WorleyJun 25, 2022, 5:50 PM
27 points
22 comments14 min readLW link

[Question] How “should” coun­ter­fac­tual pre­dic­tion mar­kets work?

eapiJun 25, 2022, 5:44 PM
9 points
6 comments1 min readLW link

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

Orpheus16Jun 25, 2022, 5:36 PM
114 points
38 comments2 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)

Some re­flec­tions on the LW com­mu­nity af­ter sev­eral months of ac­tive engagement

M. Y. ZuoJun 25, 2022, 5:04 PM
72 points
40 comments4 min readLW link

On The Spec­trum, On The Guest List: (vii) The Marquee

party girlJun 25, 2022, 4:54 PM
5 points
0 comments19 min readLW link
(onthespectrumontheguestlist.substack.com)

Iden­ti­fi­ca­tion of Nat­u­ral Modularity

Stephen FowlerJun 25, 2022, 3:05 PM
15 points
3 comments7 min readLW link

[LQ] Some Thoughts on Mes­sag­ing Around AI Risk

DragonGodJun 25, 2022, 1:53 PM
5 points
3 comments6 min readLW link

Quick Sum­maries of Two Papers on Kant and Game Theory

Erich_GrunewaldJun 25, 2022, 10:25 AM
8 points
2 comments4 min readLW link
(www.erichgrunewald.com)

[Question] Do you con­sider your cur­rent, non-su­per­hu­man self al­igned with “hu­man­ity” already?

Rana DexsinJun 25, 2022, 4:15 AM
10 points
19 comments1 min readLW link

LW/​ACX/​EA Seat­tle sum­mer meetup

nsokolskyJun 24, 2022, 11:30 PM
4 points
2 comments1 min readLW link

Depen­den­cies for AGI pessimism

YitzJun 24, 2022, 10:25 PM
7 points
4 comments1 min readLW link

[Link] Child­care : what the sci­ence says

Gunnar_ZarnckeJun 24, 2022, 9:45 PM
46 points
4 comments1 min readLW link
(criticalscience.medium.com)

What if the best path for a per­son who wants to work on AGI al­ign­ment is to join Face­book or Google?

dbaschJun 24, 2022, 9:23 PM
2 points
3 comments1 min readLW link

[Link] Ad­ver­sar­i­ally trained neu­ral rep­re­sen­ta­tions may already be as ro­bust as cor­re­spond­ing biolog­i­cal neu­ral representations

Gunnar_ZarnckeJun 24, 2022, 8:51 PM
35 points
9 comments1 min readLW link

Up­dated Defer­ence is not a strong ar­gu­ment against the util­ity un­cer­tainty ap­proach to alignment

Ivan VendrovJun 24, 2022, 7:32 PM
26 points
8 comments4 min readLW link

Cracks in the Wall, Part I: The Conscious

siloJun 24, 2022, 6:29 PM
−3 points
28 comments12 min readLW link
(stephenfoster.substack.com)

[Question] Do al­ign­ment con­cerns ex­tend to pow­er­ful non-AI agents?

OzyrusJun 24, 2022, 6:26 PM
21 points
13 comments1 min readLW link

Raphaël Millière on Gen­er­al­iza­tion and Scal­ing Maximalism

Michaël TrazziJun 24, 2022, 6:18 PM
21 points
2 comments4 min readLW link
(theinsideview.ai)

Worked Ex­am­ples of Shap­ley Values

lalaithionJun 24, 2022, 5:13 PM
75 points
11 comments8 min readLW link

Fea­ture re­quest: vot­ing but­tons at the bot­tom?

Oliver SourbutJun 24, 2022, 2:41 PM
71 points
12 comments1 min readLW link

In­tel­li­gence in Com­mit­ment Races

David UdellJun 24, 2022, 2:30 PM
28 points
8 comments5 min readLW link

Linkpost: Robin Han­son—Why Not Wait On AI Risk?

Yair HalberstadtJun 24, 2022, 2:23 PM
41 points
14 comments1 min readLW link
(www.overcomingbias.com)

[Question] “Science Cathe­drals”

Alex VermillionJun 24, 2022, 3:30 AM
22 points
9 comments1 min readLW link

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

Book re­view: The Pas­sen­ger by Lisa Lutz

KatjaGraceJun 23, 2022, 11:10 PM
12 points
1 comment1 min readLW link
(worldspiritsockpuppet.com)

20 Cri­tiques of AI Safety That I Found on Twitter

dkirmaniJun 23, 2022, 7:23 PM
21 points
16 comments1 min readLW link

The Limits of Automation

milkandcigarettesJun 23, 2022, 6:03 PM
5 points
1 comment5 min readLW link
(milkandcigarettes.com)

[Question] Is CIRL a promis­ing agenda?

Chris_LeongJun 23, 2022, 5:12 PM
28 points
16 comments1 min readLW link

[Link] OpenAI: Learn­ing to Play Minecraft with Video PreTrain­ing (VPT)

Aryeh EnglanderJun 23, 2022, 4:29 PM
53 points
3 comments1 min readLW link

Half-baked AI Safety ideas thread

Aryeh EnglanderJun 23, 2022, 4:11 PM
64 points
63 comments1 min readLW link

Non­profit Boards are Weird

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

Covid 6/​23/​22: Un­der Five Alive

ZviJun 23, 2022, 2:00 PM
26 points
9 comments10 min readLW link
(thezvi.wordpress.com)

How do states re­spond to changes in nu­clear risk

NathanBarnardJun 23, 2022, 12:42 PM
8 points
2 comments5 min readLW link

[Question] What’s the con­tin­gency plan if we get AGI to­mor­row?

YitzJun 23, 2022, 3:10 AM
61 points
23 comments1 min readLW link

[Question] What are the best “policy” ap­proaches in wor­lds where al­ign­ment is difficult?

LHAJun 23, 2022, 1:53 AM
1 point
0 comments1 min readLW link

AI Train­ing Should Allow Opt-Out

alyssavanceJun 23, 2022, 1:33 AM
76 points
13 comments6 min readLW link

Loose thoughts on AGI risk

YitzJun 23, 2022, 1:02 AM
7 points
3 comments1 min readLW link

Air Con­di­tioner Test Re­sults & Discussion

johnswentworthJun 22, 2022, 10:26 PM
82 points
42 comments6 min readLW link

An­nounc­ing the LessWrong Cu­rated Podcast

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

Google’s new text-to-image model—Parti, a demon­stra­tion of scal­ing benefits

KaydenJun 22, 2022, 8:00 PM
32 points
4 comments1 min readLW link

Build­ing an Epistemic Sta­tus Tracker

rcuJun 22, 2022, 6:57 PM
7 points
8 comments1 min readLW link

Con­fu­sion about neu­ro­science/​cog­ni­tive sci­ence as a dan­ger for AI Alignment

Samuel NellessenJun 22, 2022, 5:59 PM
3 points
1 comment3 min readLW link
(snellessen.com)

[Question] How do I use caf­feine op­ti­mally?

randomstringJun 22, 2022, 5:59 PM
18 points
31 comments1 min readLW link

Make learn­ing a reality

Dalton MaberyJun 22, 2022, 3:58 PM
13 points
2 comments1 min readLW link

Reflec­tion Mechanisms as an Align­ment tar­get: A survey

Jun 22, 2022, 3:05 PM
32 points
1 comment14 min readLW link