Dis­cuss AI Policy Recommendations

GilesApr 25, 2023, 2:21 PM
8 points
0 comments1 min readLW link

Ex­plain­ing the Trans­former Cir­cuits Frame­work by Example

Felix HofstätterApr 25, 2023, 1:45 PM
8 points
1 comment15 min readLW link

Notes on Po­ten­tial Fu­ture AI Tax Policy

ZviApr 25, 2023, 1:30 PM
33 points
6 comments9 min readLW link
(thezvi.wordpress.com)

Sen­tience in Sili­con: The Challenges of AI Consciousness

Hannes ThurnherrApr 25, 2023, 1:15 PM
5 points
2 comments5 min readLW link

Paths to failure

Apr 25, 2023, 8:03 AM
29 points
1 comment8 min readLW link

My Assess­ment of the Chi­nese AI Safety Community

Lao MeinApr 25, 2023, 4:21 AM
252 points
95 comments3 min readLW link

Mak­ing Nanobots isn’t a one-shot pro­cess, even for an ar­tifi­cial superintelligance

dankradApr 25, 2023, 12:39 AM
20 points
13 comments6 min readLW link

Men­tal Models Of Peo­ple Can Be People

Nox MLApr 25, 2023, 12:03 AM
15 points
55 comments8 min readLW link

Progress links and tweets, 2023-04-24

jasoncrawfordApr 24, 2023, 9:17 PM
16 points
1 comment2 min readLW link
(rootsofprogress.org)

Ideas for AI labs: Read­ing list

Zach Stein-PerlmanApr 24, 2023, 7:00 PM
11 points
0 comments4 min readLW link

Deep learn­ing mod­els might be se­cretly (al­most) linear

berenApr 24, 2023, 6:43 PM
117 points
29 comments4 min readLW link

Sub­jec­tive AI/​ML Digest: April II

Boris TApr 24, 2023, 6:33 PM
1 point
0 comments1 min readLW link
(borisagain.substack.com)

The Tox­o­plasma of AGI Doom and Ca­pa­bil­ities?

Robert_AIZIApr 24, 2023, 6:11 PM
72 points
12 comments1 min readLW link

[Question] Mea­sures of In­ter­net Viral­ity and News Popularity

T431Apr 24, 2023, 5:43 PM
4 points
4 comments1 min readLW link

A con­cise sum-up of the ba­sic ar­gu­ment for AI doom

Mergimio H. DoefevmilApr 24, 2023, 5:37 PM
11 points
6 comments2 min readLW link

A re­sponse to Con­jec­ture’s CoEm proposal

Kristian FreedApr 24, 2023, 5:23 PM
7 points
0 comments4 min readLW link

Ca­ma­raderie at scale: in search of shared identity

eqApr 24, 2023, 4:46 PM
8 points
2 comments8 min readLW link

A Hy­po­thet­i­cal Takeover Sce­nario Twit­ter Poll

ZviApr 24, 2023, 2:00 PM
54 points
9 comments17 min readLW link
(thezvi.wordpress.com)

Cape Town, South Africa—ACX Mee­tups Every­where “Spring” 2023

moyamoApr 24, 2023, 1:37 PM
2 points
0 comments1 min readLW link

Cred­ible, costly, pseudonymity

M. Y. ZuoApr 24, 2023, 1:35 PM
1 point
11 comments1 min readLW link

On Ar­tifice and Intelligence

Jonathan YanApr 24, 2023, 1:26 PM
2 points
0 comments1 min readLW link
(medium.com)

AGI ruin mostly rests on strong claims about al­ign­ment and de­ploy­ment, not about society

Rob BensingerApr 24, 2023, 1:06 PM
70 points
8 comments6 min readLW link

For al­ign­ment, we should si­mul­ta­neously use mul­ti­ple the­o­ries of cog­ni­tion and value

Roman LeventovApr 24, 2023, 10:37 AM
23 points
5 comments5 min readLW link

Power laws in Speedrun­ning and Ma­chine Learning

Apr 24, 2023, 10:06 AM
71 points
1 comment1 min readLW link
(arxiv.org)

[Question] “User does not meet the re­quire­ments to vote”

MonkleApr 24, 2023, 9:53 AM
4 points
3 comments1 min readLW link

The Brain is Not Close to Ther­mo­dy­namic Limits on Computation

DaemonicSigilApr 24, 2023, 8:21 AM
167 points
58 comments5 min readLW link

Value Learn­ing – Towards Re­solv­ing Con­fu­sion

PashaKamyshevApr 24, 2023, 6:43 AM
4 points
0 comments18 min readLW link

Sum­maries of top fo­rum posts (17th − 23rd April 2023)

Zoe WilliamsApr 24, 2023, 4:13 AM
18 points
0 comments8 min readLW link

Do LLMs dream of emer­gent sheep?

ShmiApr 24, 2023, 3:26 AM
16 points
2 comments1 min readLW link

Not us­ing a pri­ori in­for­ma­tion for Rus­sian propaganda

EniScienApr 24, 2023, 1:14 AM
−5 points
4 comments1 min readLW link

Con­tra Yud­kowsky on AI Doom

jacob_cannellApr 24, 2023, 12:20 AM
90 points
111 comments9 min readLW link

Con­se­quen­tial­ism is in the Stars not Ourselves

DragonGodApr 24, 2023, 12:02 AM
7 points
19 comments5 min readLW link

When did hu­mans be­come self-aware?

Derek M. JonesApr 23, 2023, 10:36 PM
6 points
2 comments1 min readLW link
(vectors.substack.com)

[Question] Are there AI poli­cies that are ro­bustly net-pos­i­tive even when con­sid­er­ing differ­ent AI sce­nar­ios?

Noosphere89Apr 23, 2023, 9:46 PM
11 points
1 comment1 min readLW link

Get­ting Started With Naturalism

LoganStrohlApr 23, 2023, 9:02 PM
69 points
4 comments11 min readLW link1 review

[Question] Why do we care about agency for al­ign­ment?

Chris_LeongApr 23, 2023, 6:10 PM
22 points
19 comments1 min readLW link

Tam­ing the Fire of Intelligence

Peter KuhnApr 23, 2023, 5:41 PM
0 points
7 comments5 min readLW link

Prevent­ing AI Mi­suse: State of the Art Re­search and its Flaws

Madhav MalhotraApr 23, 2023, 5:37 PM
15 points
0 comments11 min readLW link
(forum.effectivealtruism.org)

[Question] Could trans­former net­work mod­els learn mo­tor plan­ning like they can learn lan­guage and image gen­er­a­tion?

mu_(negative)Apr 23, 2023, 5:24 PM
2 points
4 comments1 min readLW link

Could a su­per­in­tel­li­gence de­duce gen­eral rel­a­tivity from a fal­ling ap­ple? An investigation

titotalApr 23, 2023, 12:49 PM
149 points
39 comments9 min readLW link

Endo-, Dia-, Para-, and Ecto-sys­temic novelty

TsviBTApr 23, 2023, 12:25 PM
17 points
3 comments5 min readLW link

An In­tro to An­thropic Rea­son­ing us­ing the ‘Boy or Girl Para­dox’ as a toy example

TobyCApr 23, 2023, 10:20 AM
31 points
28 comments19 min readLW link

[Question] Se­man­tics, Syn­tax and Prag­mat­ics of the Mind?

Ben AmitayApr 23, 2023, 6:13 AM
2 points
0 comments1 min readLW link

A great talk for AI noobs (ac­cord­ing to an AI noob)

dovApr 23, 2023, 5:34 AM
10 points
1 comment1 min readLW link
(forum.effectivealtruism.org)

Bits of NEFFA

jefftkApr 23, 2023, 2:20 AM
5 points
0 comments1 min readLW link
(www.jefftk.com)

“Rate limit­ing” as a mod tool

RaemonApr 23, 2023, 12:42 AM
48 points
36 comments4 min readLW link

What should we cen­sor from train­ing data?

wassnameApr 22, 2023, 11:33 PM
16 points
4 comments1 min readLW link

Ar­chi­tec­ture-aware op­ti­mi­sa­tion: train ImageNet and more with­out hyperparameters

Chris MingardApr 22, 2023, 9:50 PM
6 points
2 comments2 min readLW link

OpenAI’s GPT-4 Safety Goals

PeterMcCluskeyApr 22, 2023, 7:11 PM
3 points
3 comments4 min readLW link
(bayesianinvestor.com)

In­tro­duc­ing the Nuts and Bolts Of Naturalism

LoganStrohlApr 22, 2023, 6:31 PM
78 points
2 comments3 min readLW link