The Sinews of Su­dan’s Lat­est War

Tim LiptrotAug 4, 2023, 6:17 PM
43 points
12 comments12 min readLW link

Pri­vate notes on LW?

RaemonAug 4, 2023, 5:35 PM
61 points
33 comments1 min readLW link

When train­ing AI, we should es­ca­late the fre­quency of ca­pa­bil­ity tests

Hauke HillebrandtAug 4, 2023, 4:07 PM
2 points
0 comments1 min readLW link

Man­i­fund: What we’re fund­ing (weeks 2-4)

Austin ChenAug 4, 2023, 4:00 PM
44 points
2 commentsLW link
(manifund.substack.com)

[Linkpost] Mul­ti­modal Neu­rons in Pre­trained Text-Only Transformers

Bogdan Ionut CirsteaAug 4, 2023, 3:29 PM
11 points
0 comments1 min readLW link

Apollo Re­search is hiring evals and in­ter­pretabil­ity en­g­ineers & scientists

Marius HobbhahnAug 4, 2023, 10:54 AM
25 points
0 comments2 min readLW link

[Question] Has any­one tried cre­at­ing a YouTube or TikTok se­ries cov­er­ing the se­quences?

Max RossiAug 4, 2023, 12:10 AM
4 points
4 comments1 min readLW link

[Question] Is there any met­ric mea­sur­ing ~”pro­por­tion of peo­ple cre­at­ing ex­tra value”?

Amal Aug 3, 2023, 10:54 PM
7 points
3 comments1 min readLW link

[Question] Hy­po­thet­i­cal: what would you do?

JNSAug 3, 2023, 10:39 PM
4 points
2 comments1 min readLW link

[Linkpost] De­cep­tion Abil­ities Emerged in Large Lan­guage Models

Bogdan Ionut CirsteaAug 3, 2023, 5:28 PM
12 points
0 comments1 min readLW link

Embed­ding Eth­i­cal Pri­ors into AI Sys­tems: A Bayesian Approach

JustausernameAug 3, 2023, 3:31 PM
−5 points
3 comments21 min readLW link

Pass­word-locked mod­els: a stress case for ca­pa­bil­ities evaluation

Fabien RogerAug 3, 2023, 2:53 PM
156 points
14 comments6 min readLW link

AI #23: Fun­da­men­tal Prob­lems with RLHF

ZviAug 3, 2023, 12:50 PM
59 points
9 comments41 min readLW link
(thezvi.wordpress.com)

Bad Imi­ta­tion Instruments

jefftkAug 3, 2023, 2:30 AM
21 points
1 comment1 min readLW link
(www.jefftk.com)

Kol­mogorov’s the­ory of Al­gorith­mic Probability

Aidan RockeAug 3, 2023, 12:58 AM
5 points
2 comments2 min readLW link
(keplerlounge.com)

Work cul­ture creep

CrimsonChinAug 3, 2023, 12:38 AM
32 points
15 comments8 min readLW link

[Question] Boxing

Zach Stein-PerlmanAug 2, 2023, 11:38 PM
6 points
1 comment1 min readLW link

Ex­ter­nal ra­tio­nal­ity vs. in­ter­nal rationality

metachiralityAug 2, 2023, 11:29 PM
7 points
0 comments1 min readLW link

When perform­ing a di­men­sion­al­ity re­duc­tion on ten­sors, the trace is of­ten zero.

Joseph Van NameAug 2, 2023, 9:06 PM
7 points
1 comment3 min readLW link

Progress links di­gest, 2023-08-02: Su­per­con­duc­tor edition

jasoncrawfordAug 2, 2023, 8:27 PM
13 points
0 comments3 min readLW link
(rootsofprogress.org)

[Question] What works for ADHD and/​or re­lated things?

TeaTieAndHatAug 2, 2023, 6:37 PM
7 points
13 comments1 min readLW link

[Question] Would you pay for a search en­g­ine limited to ra­tio­nal­ist sites?

ConorAug 2, 2023, 6:06 PM
4 points
19 comments1 min readLW link

The Roots of Progress Blog-Build­ing In­ten­sive: ad­vice for ap­pli­cants, re­quest for support

jasoncrawfordAug 2, 2023, 3:37 PM
9 points
0 comments1 min readLW link
(rootsofprogress.org)

3 lev­els of threat obfuscation

HoldenKarnofskyAug 2, 2023, 2:58 PM
69 points
14 comments7 min readLW link

ChatGPT for trans­la­tion

Varshul GuptaAug 2, 2023, 11:57 AM
1 point
0 comments3 min readLW link
(dubverseblack.substack.com)

Long-Term Fu­ture Fund: April 2023 grant recommendations

Aug 2, 2023, 7:54 AM
81 points
3 comments50 min readLW link

[Question] Could we breed/​en­g­ineer in­tel­li­gent par­rots?

lemonhopeAug 2, 2023, 7:32 AM
9 points
18 comments1 min readLW link

An­throp­i­cal Motte and Bailey in two ver­sions of Sleep­ing Beauty

Ape in the coatAug 2, 2023, 7:08 AM
32 points
56 comments6 min readLW link

so­lar-ther­mal and techno-eco­nomic analysis

bhauthAug 2, 2023, 6:22 AM
21 points
8 comments5 min readLW link
(www.bhauth.com)

South Bay ACX/​SSC Meetup @ Whole Foods

allisonaAug 2, 2023, 3:44 AM
1 point
0 comments1 min readLW link

“Is There Any­thing That’s Worth More”

Zack_M_DavisAug 2, 2023, 3:28 AM
64 points
6 comments1 min readLW link

Bay Win­ter Sols­tice: call for speech pitches!

tcheasdfjklAug 2, 2023, 3:24 AM
9 points
0 comments1 min readLW link
(docs.google.com)

[Question] What is on­tol­ogy?

Adam ZernerAug 2, 2023, 12:54 AM
28 points
19 comments1 min readLW link

My cur­rent LK99 questions

Eliezer YudkowskyAug 1, 2023, 10:48 PM
206 points
38 comments5 min readLW link

Spiral Staircase

Michael SamoilovAug 1, 2023, 9:51 PM
19 points
2 comments2 min readLW link

Open Mic—Au­gust 2023

Adam ZernerAug 1, 2023, 7:24 PM
8 points
0 comments1 min readLW link

ARC Evals new re­port: Eval­u­at­ing Lan­guage-Model Agents on Real­is­tic Au­tonomous Tasks

Beth BarnesAug 1, 2023, 6:30 PM
153 points
12 comments5 min readLW link
(evals.alignment.org)

[Question] When(if ever) are su­per­stim­uli good/​use­ful/​ad­van­ta­geous?

PerhapsAug 1, 2023, 3:50 PM
−7 points
2 comments1 min readLW link

AISN #17: Au­to­mat­i­cally Cir­cum­vent­ing LLM Guardrails, the Fron­tier Model Fo­rum, and Se­nate Hear­ing on AI Oversight

Dan HAug 1, 2023, 3:40 PM
8 points
0 comments8 min readLW link
(newsletter.safe.ai)

AISN #16: White House Se­cures Vol­un­tary Com­mit­ments from Lead­ing AI Labs and Les­sons from Oppenheimer

Aug 1, 2023, 3:39 PM
3 points
0 comments6 min readLW link
(newsletter.safe.ai)

“Des­per­ate Hon­esty” by Agnes Callard

David GrossAug 1, 2023, 1:34 PM
11 points
0 comments2 min readLW link
(dailynous.com)

Bar­bieheimer: Across the Dead Reckoning

ZviAug 1, 2023, 1:00 PM
49 points
17 comments41 min readLW link
(thezvi.wordpress.com)

Un­tan­gling In­frabayesi­anism: A re­dis­til­la­tion [PDF link; ~12k words + lots of math]

LorxusAug 1, 2023, 12:42 PM
29 points
16 comments2 min readLW link
(docdro.id)

What Is Child­hood Sup­posed To Be?

SableAug 1, 2023, 9:51 AM
21 points
13 comments3 min readLW link
(affablyevil.substack.com)

AI ro­man­tic part­ners will harm so­ciety if they go unregulated

Roman LeventovAug 1, 2023, 9:32 AM
26 points
76 comments13 min readLW link

What is au­ton­omy, and how does it lead to greater risk from AI?

DavidmanheimAug 1, 2023, 7:58 AM
30 points
0 comments6 min readLW link

Eval­u­at­ing Su­per­hu­man Models with Con­sis­tency Checks

Aug 1, 2023, 7:51 AM
21 points
2 comments9 min readLW link
(arxiv.org)

[See link to Sept meetup be­low!] San Fran­cisco ACX Meetup “First Satur­day” Au­gust 5, 1 pm

guenaelAug 1, 2023, 3:38 AM
1 point
0 comments1 min readLW link

[Question] Ex­er­cise: Solve “Think­ing Physics”

RaemonAug 1, 2023, 12:44 AM
102 points
30 comments5 min readLW link1 review

The “pub­lic de­bate” about AI is con­fus­ing for the gen­eral pub­lic and for poli­cy­mak­ers be­cause it is a three-sided de­bate

Adam David LongAug 1, 2023, 12:08 AM
146 points
30 comments4 min readLW link