Do you work at an AI lab? Please quit

Nik SamoylovMay 5, 2023, 11:41 PM
−29 points
9 comments1 min readLW link

Ex­plain­ing “Hell is Game The­ory Folk The­o­rems”

electroswingMay 5, 2023, 11:33 PM
57 points
21 comments5 min readLW link

Sleep­ing Beauty – the Death Hypothesis

Guillaume CharrierMay 5, 2023, 11:32 PM
7 points
8 comments5 min readLW link

Orthog­o­nal’s For­mal-Goal Align­ment the­ory of change

Tamsin LeakeMay 5, 2023, 10:36 PM
69 points
13 comments4 min readLW link
(carado.moe)

A smart enough LLM might be deadly sim­ply if you run it for long enough

Mikhail SaminMay 5, 2023, 8:49 PM
19 points
16 comments8 min readLW link

What Ja­son has been read­ing, May 2023: “Pro­topia,” com­plex sys­tems, Daedalus vs. Icarus, and more

jasoncrawfordMay 5, 2023, 7:54 PM
25 points
2 comments11 min readLW link
(rootsofprogress.org)

CHAT Di­plo­macy: LLMs and Na­tional Security

SebastianG May 5, 2023, 7:45 PM
25 points
6 comments7 min readLW link

Linkpost for Ac­cursed Farms Dis­cus­sion /​ de­bate with AI ex­pert Eliezer Yudkowsky

gilchMay 5, 2023, 6:20 PM
14 points
2 comments1 min readLW link
(www.youtube.com)

Reg­u­late or Com­pete? The China Fac­tor in U.S. AI Policy (NAIR #2)

charles_mMay 5, 2023, 5:43 PM
2 points
1 comment7 min readLW link
(navigatingairisks.substack.com)

Kingfisher Live CD Process

jefftkMay 5, 2023, 5:00 PM
13 points
0 comments3 min readLW link
(www.jefftk.com)

What can we learn from Bayes about rea­son­ing?

jasoncrawfordMay 5, 2023, 3:52 PM
22 points
11 comments1 min readLW link

[Question] Why not use ac­tive SETI to pre­vent AI Doom?

RomanSMay 5, 2023, 2:41 PM
13 points
13 comments1 min readLW link

In­ves­ti­gat­ing Emer­gent Goal-Like Be­hav­ior in Large Lan­guage Models us­ing Ex­per­i­men­tal Economics

phelps-sgMay 5, 2023, 11:15 AM
6 points
1 comment4 min readLW link

Monthly Shorts 4/​23

CelerMay 5, 2023, 7:20 AM
8 points
1 comment3 min readLW link
(keller.substack.com)

[Question] What is it like to be a com­pat­i­bil­ist?

tslarmMay 5, 2023, 2:56 AM
8 points
72 comments1 min readLW link

Tran­script of a pre­sen­ta­tion on catas­trophic risks from AI

RobertMMay 5, 2023, 1:38 AM
6 points
0 comments8 min readLW link

How to get good at programming

Ulisse MiniMay 5, 2023, 1:14 AM
39 points
3 comments2 min readLW link

An Up­date On The Cam­paign For AI Safety Dot Org

yanni kyriacosMay 5, 2023, 12:21 AM
−13 points
2 comments1 min readLW link

A brief col­lec­tion of Hin­ton’s re­cent com­ments on AGI risk

Kaj_SotalaMay 4, 2023, 11:31 PM
143 points
9 comments11 min readLW link

Robin Han­son and I talk about AI risk

KatjaGraceMay 4, 2023, 10:20 PM
39 points
8 comments1 min readLW link
(worldspiritsockpuppet.com)

Who reg­u­lates the reg­u­la­tors? We need to go be­yond the re­view-and-ap­proval paradigm

jasoncrawfordMay 4, 2023, 10:11 PM
122 points
29 comments13 min readLW link
(rootsofprogress.org)

Re­cur­sive Mid­dle Man­ager Hell: AI Edition

VojtaKovarikMay 4, 2023, 8:08 PM
30 points
11 comments2 min readLW link

AI risk/​re­ward: A sim­ple model

Nathan YoungMay 4, 2023, 7:25 PM
3 points
0 commentsLW link

Google “We Have No Moat, And Nei­ther Does OpenAI”

Chris_LeongMay 4, 2023, 6:23 PM
61 points
28 comments1 min readLW link
(www.semianalysis.com)

Try­ing to mea­sure AI de­cep­tion ca­pa­bil­ities us­ing tem­po­rary simu­la­tion fine-tuning

alenoachMay 4, 2023, 5:59 PM
4 points
0 comments7 min readLW link

[Linkpost]Trans­former-Based LM Sur­prisal Pre­dicts Hu­man Read­ing Times Best with About Two Billion Train­ing Tokens

Curtis HuebnerMay 4, 2023, 5:16 PM
10 points
1 comment1 min readLW link
(arxiv.org)

Clar­ify­ing and pre­dict­ing AGI

Richard_NgoMay 4, 2023, 3:55 PM
142 points
45 comments4 min readLW link

[Cross­post] AI X-risk in the News: How Effec­tive are Re­cent Me­dia Items and How is Aware­ness Chang­ing? Our New Sur­vey Re­sults.

otto.bartenMay 4, 2023, 2:09 PM
5 points
0 comments9 min readLW link
(forum.effectivealtruism.org)

AI #10: Code In­ter­preter and Ge­off Hinton

ZviMay 4, 2023, 2:00 PM
80 points
7 comments78 min readLW link
(thezvi.wordpress.com)

Ad­vice for in­ter­act­ing with busy people

Severin T. SeehrichMay 4, 2023, 1:31 PM
68 points
4 comments4 min readLW link

We don’t need AGI for an amaz­ing future

Karl von WendtMay 4, 2023, 12:10 PM
19 points
32 comments5 min readLW link

Has the Sym­bol Ground­ing Prob­lem just gone away?

RussellThorMay 4, 2023, 7:46 AM
12 points
3 comments1 min readLW link

Opinion merg­ing for AI control

David JohnstonMay 4, 2023, 2:43 AM
6 points
0 comments11 min readLW link

Un­der­stand­ing why illu­sion­ism does not deny the ex­is­tence of qualia

Mergimio H. DoefevmilMay 4, 2023, 2:13 AM
0 points
17 comments1 min readLW link

[New] Re­jected Con­tent Section

May 4, 2023, 1:43 AM
65 points
21 comments5 min readLW link

How MATS ad­dresses “mass move­ment build­ing” concerns

Ryan KiddMay 4, 2023, 12:55 AM
63 points
9 comments3 min readLW link

Mov­ing VPS Again

jefftkMay 4, 2023, 12:30 AM
9 points
2 comments1 min readLW link
(www.jefftk.com)

Prizes for ma­trix com­ple­tion problems

paulfchristianoMay 3, 2023, 11:30 PM
164 points
52 comments1 min readLW link
(www.alignment.org)

Align­ment Re­search @ EleutherAI

Curtis HuebnerMay 3, 2023, 10:45 PM
40 points
1 comment3 min readLW link
(blog.eleuther.ai)

«Boundaries/​Mem­branes» and AI safety compilation

ChipmonkMay 3, 2023, 9:41 PM
56 points
17 comments8 min readLW link

[Question] What con­straints does deep learn­ing place on al­ign­ment plans?

Garrett BakerMay 3, 2023, 8:40 PM
9 points
0 comments1 min readLW link

AGI ris­ing: why we are in a new era of acute risk and in­creas­ing pub­lic aware­ness, and what to do now

Greg CMay 3, 2023, 8:26 PM
25 points
12 commentsLW link

For­mal­iz­ing the “AI x-risk is un­likely be­cause it is ridicu­lous” argument

Christopher KingMay 3, 2023, 6:56 PM
48 points
17 comments3 min readLW link

[Question] List of no­table peo­ple who be­lieve in AI X-risk?

vlad.proexMay 3, 2023, 6:46 PM
14 points
4 comments1 min readLW link

[Question] LessWrong ex­port­ing?

axiomAdministratorMay 3, 2023, 6:34 PM
0 points
3 comments1 min readLW link

Progress links and tweets, 2023-05-03

jasoncrawfordMay 3, 2023, 4:23 PM
13 points
0 comments2 min readLW link
(rootsofprogress.org)

Per­son­hood is a Reli­gious Belief

jan SijanMay 3, 2023, 4:16 PM
−41 points
28 comments6 min readLW link

Slow­ing AI: Crunch time

Zach Stein-PerlmanMay 3, 2023, 3:00 PM
11 points
1 comment2 min readLW link

Find­ing Neu­rons in a Haystack: Case Stud­ies with Sparse Probing

May 3, 2023, 1:30 PM
33 points
6 comments2 min readLW link1 review
(arxiv.org)

Monthly Roundup #6: May 2023

ZviMay 3, 2023, 12:50 PM
31 points
12 comments24 min readLW link
(thezvi.wordpress.com)