AI x-risk, ap­prox­i­mately or­dered by embarrassment

Alex Lawsen Apr 12, 2023, 11:01 PM
151 points
7 comments19 min readLW link

AXRP Epi­sode 20 - ‘Re­form’ AI Align­ment with Scott Aaronson

DanielFilanApr 12, 2023, 9:30 PM
22 points
2 comments68 min readLW link

Ap­ply to >30 AI safety fun­ders in one ap­pli­ca­tion with the Non­lin­ear Network

Apr 12, 2023, 9:23 PM
65 points
12 comments2 min readLW link

AGI goal space is big, but nar­row­ing might not be as hard as it seems.

Jacy Reese AnthisApr 12, 2023, 7:03 PM
15 points
0 comments3 min readLW link

Nat­u­ral lan­guage alignment

Jacy Reese AnthisApr 12, 2023, 7:02 PM
31 points
2 comments2 min readLW link

Repug­nant lev­els of violins

Solenoid_EntityApr 12, 2023, 5:11 PM
73 points
10 comments12 min readLW link

Progress links and tweets, 2023-04-12

jasoncrawfordApr 12, 2023, 4:52 PM
8 points
2 comments1 min readLW link
(rootsofprogress.org)

A ba­sic math­e­mat­i­cal struc­ture of intelligence

GololApr 12, 2023, 4:49 PM
4 points
6 comments4 min readLW link

[Question] Should Au­toGPT up­date us to­wards re­search­ing IDA?

Michaël TrazziApr 12, 2023, 4:41 PM
15 points
5 comments1 min readLW link

Box­ing lessons

yakimoffApr 12, 2023, 4:19 PM
1 point
0 comments1 min readLW link

Dazed and con­fused: Good olde’ walk around the Marin Head­lands

yakimoffApr 12, 2023, 4:09 PM
1 point
0 comments1 min readLW link

Towards a solu­tion to the al­ign­ment prob­lem via ob­jec­tive de­tec­tion and eval­u­a­tion

Paul CologneseApr 12, 2023, 3:39 PM
9 points
7 comments12 min readLW link

Ar­tifi­cial In­tel­li­gence as exit strat­egy from the age of acute ex­is­ten­tial risk

Arturo MaciasApr 12, 2023, 2:48 PM
−7 points
15 comments7 min readLW link

The UBI dystopia: a glimpse into the fu­ture via pre­sent-day abuses

Solenoid_EntityApr 12, 2023, 2:44 PM
50 points
73 comments4 min readLW link

[Question] Goals of model vs. goals of simu­lacra?

dr_sApr 12, 2023, 1:02 PM
5 points
7 comments1 min readLW link

Align­ment of Au­toGPT agents

OzyrusApr 12, 2023, 12:54 PM
14 points
1 comment4 min readLW link

Boundaries-based se­cu­rity and AI safety approaches

Allison DuettmannApr 12, 2023, 12:36 PM
43 points
2 comments6 min readLW link

Scaf­folded LLMs as nat­u­ral lan­guage computers

berenApr 12, 2023, 10:47 AM
95 points
10 comments11 min readLW link

LW is prob­a­bly not the place for “I asked this LLM (x) and here’s what it said!”, but where is?

lillybaeumApr 12, 2023, 10:12 AM
21 points
3 comments1 min readLW link

No con­vinc­ing ev­i­dence for gra­di­ent de­scent in ac­ti­va­tion space

BlaineApr 12, 2023, 4:48 AM
85 points
9 comments20 min readLW link

A Brief In­tro­duc­tion to ACI, 2: An Event-Cen­tric View

Akira PyinyaApr 12, 2023, 3:23 AM
3 points
0 comments2 min readLW link

Bos­ton So­cial Dance Covid Requirements

jefftkApr 12, 2023, 2:30 AM
7 points
2 comments1 min readLW link
(www.jefftk.com)

[Link] Sarah Con­stantin: “Why I am Not An AI Doomer”

lbThingrbApr 12, 2023, 1:52 AM
61 points
13 comments1 min readLW link
(sarahconstantin.substack.com)

[Question] Ra­tion­al­ist po­si­tion to­wards ly­ing?

WilliamTerryApr 12, 2023, 1:21 AM
−2 points
4 comments1 min readLW link

Na­tional Telecom­mu­ni­ca­tions and In­for­ma­tion Ad­minis­tra­tion: AI Ac­countabil­ity Policy Re­quest for Comment

sanxiynApr 11, 2023, 10:59 PM
9 points
0 comments1 min readLW link
(ntia.gov)

Bi­nar­is­tic Bifur­ca­tion: How Real­ity Splits Into Two Separate Binaries

Thoth HermesApr 11, 2023, 9:19 PM
−25 points
0 comments3 min readLW link
(thothhermes.substack.com)

Bryan Bishop AMA on the Progress Forum

jasoncrawfordApr 11, 2023, 9:05 PM
8 points
0 comments1 min readLW link
(rootsofprogress.org)

AI Risk US Pres­i­den­tial Candidate

Simon BerensApr 11, 2023, 7:31 PM
5 points
3 comments1 min readLW link

Evolu­tion pro­vides no ev­i­dence for the sharp left turn

Quintin PopeApr 11, 2023, 6:43 PM
206 points
65 comments15 min readLW link1 review

On “aiming for con­ver­gence on truth”

gjmApr 11, 2023, 6:19 PM
67 points
55 comments13 min readLW link

In fa­vor of ac­cel­er­at­ing prob­lems you’re try­ing to solve

Christopher KingApr 11, 2023, 6:15 PM
2 points
2 comments4 min readLW link

[In­ter­view w/​ Jeffrey Ladish] Ap­ply­ing the ‘se­cu­rity mind­set’ to AI and x-risk

fowlertmApr 11, 2023, 6:14 PM
12 points
0 comments1 min readLW link

Re­quest to AGI or­ga­ni­za­tions: Share your views on paus­ing AI progress

Apr 11, 2023, 5:30 PM
141 points
11 comments1 min readLW link

Core of AI pro­jec­tions from first prin­ci­ples: At­tempt 1

tailcalledApr 11, 2023, 5:24 PM
21 points
3 comments3 min readLW link

What Ja­son has been read­ing, April 2023

jasoncrawfordApr 11, 2023, 4:29 PM
18 points
0 comments5 min readLW link
(rootsofprogress.org)

What about an AI that’s SUPPOSED to kill us (not ChaosGPT; only on pa­per)?

False NameApr 11, 2023, 4:09 PM
−13 points
1 comment3 min readLW link

Con­tra-Berkeley

False NameApr 11, 2023, 4:06 PM
0 points
0 comments4 min readLW link

Con­tra-Wittgen­stein; no postmodernism

False NameApr 11, 2023, 4:05 PM
−16 points
1 comment5 min readLW link

[MLSN #9] Ver­ify­ing large train­ing runs, se­cu­rity risks from LLM ac­cess to APIs, why nat­u­ral se­lec­tion may fa­vor AIs over humans

Apr 11, 2023, 4:03 PM
11 points
0 comments6 min readLW link
(newsletter.mlsafety.org)

Where’s the foom?

Fergus FettesApr 11, 2023, 3:50 PM
34 points
27 comments2 min readLW link

“The Need for Long-term Re­search”—Seeds of Science call for reviewers

rogersbaconApr 11, 2023, 3:37 PM
15 points
0 comments1 min readLW link

NTIA—AI Ac­countabil­ity Announcement

samshapApr 11, 2023, 3:03 PM
7 points
0 comments1 min readLW link
(www.ntia.doc.gov)

A cou­ple of ques­tions about Con­jec­ture’s Cog­ni­tive Emu­la­tion proposal

Igor IvanovApr 11, 2023, 2:05 PM
30 points
1 comment3 min readLW link

Child­hood Roundup #2

ZviApr 11, 2023, 1:50 PM
31 points
4 comments19 min readLW link
(thezvi.wordpress.com)

Mea­sur­ing ar­tifi­cial in­tel­li­gence on hu­man bench­marks is naive

AnomalousApr 11, 2023, 11:34 AM
11 points
4 comments1 min readLW link
(forum.effectivealtruism.org)

Killing Socrates

Duncan Sabien (Inactive)Apr 11, 2023, 10:28 AM
196 points
146 comments8 min readLW link1 review

Cy­berspace Ad­minis­tra­tion of China: Draft of “Reg­u­la­tion for Gen­er­a­tive Ar­tifi­cial In­tel­li­gence Ser­vices” is open for comments

sanxiynApr 11, 2023, 9:32 AM
7 points
2 comments1 min readLW link
(archive.is)

[Question] Is “Strong Co­her­ence” Anti-Nat­u­ral?

DragonGodApr 11, 2023, 6:22 AM
23 points
25 comments2 min readLW link

Four mind­set dis­agree­ments be­hind ex­is­ten­tial risk dis­agree­ments in ML

Rob BensingerApr 11, 2023, 4:53 AM
137 points
12 commentsLW link

Align­ment vs capabilities

Adam ZernerApr 11, 2023, 4:35 AM
13 points
2 comments4 min readLW link