An­thropic is fur­ther ac­cel­er­at­ing the Arms Race?

sapphireApr 6, 2023, 11:29 PM
82 points
22 comments1 min readLW link
(techcrunch.com)

Sugges­tion for safe AI struc­ture (Cu­rated Trans­par­ent De­ci­sions)

Kane GregoryApr 6, 2023, 10:00 PM
5 points
6 comments3 min readLW link

10 rea­sons why lists of 10 rea­sons might be a win­ning strategy

trevorApr 6, 2023, 9:24 PM
110 points
7 comments1 min readLW link

A Defense of Utilitarianism

Pareto OptimalApr 6, 2023, 9:09 PM
−3 points
2 comments5 min readLW link
(paretooptimal.substack.com)

One Does Not Sim­ply Re­place the Hu­mans

JerkyTreatsApr 6, 2023, 8:56 PM
9 points
3 comments4 min readLW link
(www.lesswrong.com)

[Question] Where to be­gin in ML/​AI?

Jake the StudentApr 6, 2023, 8:45 PM
9 points
4 comments1 min readLW link

Mis­gen­er­al­iza­tion as a misnomer

So8resApr 6, 2023, 8:43 PM
129 points
22 comments4 min readLW link

You can use GPT-4 to cre­ate prompt in­jec­tions against GPT-4

WitchBOTApr 6, 2023, 8:39 PM
87 points
8 comments2 min readLW link

AI scares and chang­ing pub­lic beliefs

Seth HerdApr 6, 2023, 6:51 PM
46 points
21 comments6 min readLW link

AISafety.world is a map of the AIS ecosystem

Hamish DoodlesApr 6, 2023, 6:37 PM
80 points
0 comments1 min readLW link

I asked my sen­a­tor to slow AI

OmidApr 6, 2023, 6:18 PM
21 points
5 comments2 min readLW link

Pause AI Devel­op­ment?

PeterMcCluskeyApr 6, 2023, 5:23 PM
11 points
0 comments2 min readLW link
(bayesianinvestor.com)

Use these three heuris­tic im­per­a­tives to solve alignment

GApr 6, 2023, 4:20 PM
−17 points
4 comments1 min readLW link

Eliezer on The Lu­nar So­ciety podcast

Max HApr 6, 2023, 4:18 PM
40 points
5 comments1 min readLW link
(www.dwarkeshpatel.com)

Do we get bet­ter or worse at adapt­ing to change?

jasoncrawfordApr 6, 2023, 2:42 PM
12 points
2 comments3 min readLW link
(rootsofprogress.org)

Is it true that only a chat­bot en­couraged a man to com­mit suicide?

Jeroen De RyckApr 6, 2023, 2:10 PM
6 points
0 comments4 min readLW link
(www.vrt.be)

A Fresh FAQ on GiveWiki and Im­pact Mar­kets Generally

Dawn DrescherApr 6, 2023, 2:02 PM
−1 points
0 commentsLW link
(impactmarkets.substack.com)

AI #6: Agents of Change

ZviApr 6, 2023, 2:00 PM
79 points
13 comments47 min readLW link
(thezvi.wordpress.com)

Stupid Ques­tions—April 2023

ChristianKlApr 6, 2023, 1:07 PM
17 points
46 comments1 min readLW link

(Yet Another) Map for AI Risk Discussion

chronolitusApr 6, 2023, 11:55 AM
1 point
0 comments2 min readLW link

The Com­pu­ta­tional Anatomy of Hu­man Values

berenApr 6, 2023, 10:33 AM
74 points
30 comments30 min readLW link

[Question] Is “Re­cur­sive Self-Im­prove­ment” Rele­vant in the Deep Learn­ing Paradigm?

DragonGodApr 6, 2023, 7:13 AM
32 points
36 comments7 min readLW link

Re­vis­it­ing the Hori­zon Length Hypothesis

Pablo VillalobosApr 6, 2023, 6:39 AM
23 points
4 comments3 min readLW link

Monthly Shorts 3/​23

CelerApr 6, 2023, 6:20 AM
7 points
1 comment4 min readLW link
(keller.substack.com)

Dual-Use­ness is a Ratio

jimrandomhApr 6, 2023, 5:46 AM
35 points
2 comments1 min readLW link

[Question] What’s the deal with Effec­tive Ac­cel­er­a­tionism (e/​acc)?

RomanHaukssonApr 6, 2023, 4:03 AM
23 points
9 comments2 min readLW link

No Sum­mer Har­vest: Why AI Devel­op­ment Won’t Pause

Stephen FowlerApr 6, 2023, 3:53 AM
14 points
17 comments12 min readLW link

Yoshua Ben­gio: “Slow­ing down de­vel­op­ment of AI sys­tems pass­ing the Tur­ing test”

Roman LeventovApr 6, 2023, 3:31 AM
49 points
2 comments5 min readLW link
(yoshuabengio.org)

Unal­igned sta­ble loops emerge at scale

Michael TontchevApr 6, 2023, 2:15 AM
9 points
8 comments4 min readLW link

Some­one already tried “Chaos-GPT”

robert-croninApr 6, 2023, 2:15 AM
17 points
4 comments1 min readLW link

[Question] Daisy-chain­ing ep­silon-step verifiers

DecaeneusApr 6, 2023, 2:07 AM
2 points
1 comment1 min readLW link

Auto-GPT: Open-sourced dis­aster?

awgApr 5, 2023, 10:46 PM
23 points
18 comments1 min readLW link
(github.com)

The Orthog­o­nal­ity Th­e­sis is Not Ob­vi­ously True

omnizoidApr 5, 2023, 9:06 PM
3 points
80 comments9 min readLW link

Willi­ams-Beuren Syn­drome: Frendly Mutations

TakkApr 5, 2023, 8:59 PM
−1 points
1 comment1 min readLW link

OpenAI: Our ap­proach to AI safety

Jacob G-WApr 5, 2023, 8:26 PM
1 point
1 comment1 min readLW link
(openai.com)

Why Are Max­i­mum En­tropy Distri­bu­tions So Ubiquitous?

johnswentworthApr 5, 2023, 8:12 PM
68 points
6 comments9 min readLW link

“On Liv­ing in an Atomic Age”, by C.S. Lewis (1948)

tjaffeeApr 5, 2023, 6:34 PM
17 points
3 comments8 min readLW link
(hebrew-streams.org)

Eliezer Yud­kowsky’s Let­ter in Time Magazine

ZviApr 5, 2023, 6:00 PM
214 points
86 comments14 min readLW link
(thezvi.wordpress.com)

Dark Ar­tifi­cial Intelligence

FrankAIApr 5, 2023, 5:37 PM
0 points
0 comments4 min readLW link

[Question] Best ar­gu­ments against in­stru­men­tal con­ver­gence?

lfrymireApr 5, 2023, 5:06 PM
5 points
7 comments1 min readLW link

Progress links and tweets, 2023-04-05

jasoncrawfordApr 5, 2023, 4:18 PM
20 points
0 comments2 min readLW link
(rootsofprogress.org)

Univer­sal­ity and Hid­den In­for­ma­tion in Con­cept Bot­tle­neck Models

HoagyApr 5, 2023, 2:00 PM
23 points
0 comments11 min readLW link

AI safety and the se­cu­rity mind­set: user in­ter­face de­sign, red-teams, for­mal verification

Allison DuettmannApr 5, 2023, 11:33 AM
35 points
0 comments8 min readLW link

ICA Simulacra

OzyrusApr 5, 2023, 6:41 AM
26 points
2 comments7 min readLW link

AGI de­ploy­ment as an act of aggression

dr_sApr 5, 2023, 6:39 AM
28 points
30 comments13 min readLW link

A Brief In­tro­duc­tion to Al­gorith­mic Com­mon In­tel­li­gence, ACI . 1

Akira PyinyaApr 5, 2023, 5:43 AM
−2 points
1 comment2 min readLW link

46% of US adults at least “some­what con­cerned” about AI ex­tinc­tion risk.

FoyleApr 5, 2023, 5:25 AM
1 point
0 comments1 min readLW link

[Question] Has any­one thought about how to pro­ceed now that AI notkil­lev­ery­oneism is be­com­ing more rele­vant/​is ap­proach­ing the Over­ton win­dow?

metachiralityApr 5, 2023, 3:06 AM
11 points
8 comments1 min readLW link

Em­pa­thy bandaid for im­me­di­ate AI catastrophe

installgentooApr 5, 2023, 2:12 AM
1 point
2 comments1 min readLW link

“Cor­rigi­bil­ity at some small length” by dath ilan

Christopher KingApr 5, 2023, 1:47 AM
32 points
3 comments9 min readLW link
(www.glowfic.com)