Near San Fran­cisco—Hike to watch the Blue Angels

PeterMcCluskeySep 16, 2023, 10:00 PM
9 points
0 comments1 min readLW link

Austin Petrov Day re­vi­sion, 2023

Austin LW ritual working groupSep 16, 2023, 9:55 PM
3 points
0 comments1 min readLW link

Ex­plor­ing Nat­u­ral Disaster Forecasting

GeoVaneSep 16, 2023, 5:01 PM
1 point
0 comments5 min readLW link

The com­ment­ing re­stric­tions on LessWrong seem bad

Bentham's BulldogSep 16, 2023, 4:38 PM
20 points
37 comments1 min readLW link

Po­lariza­tion is Not (Stan­dard) Bayesian

Kevin DorstSep 16, 2023, 4:31 PM
11 points
6 comments7 min readLW link
(kevindorst.substack.com)

[Question] Do you know of any re­li­able DIY com­pendium of home phys­i­cal ther­apy ex­er­cises?

David GrossSep 16, 2023, 2:37 PM
9 points
4 comments1 min readLW link

[Question] In the Short-Term, Why Couldn’t You Just RLHF-out In­stru­men­tal Con­ver­gence?

simeon_cSep 16, 2023, 10:44 AM
21 points
6 comments1 min readLW link

<|end­of­text|> is a van­ish­ing text?

MiguelDevSep 16, 2023, 2:34 AM
10 points
0 comments1 min readLW link

[Question] What’s up with psy­cho­net­ics?

metachiralitySep 16, 2023, 1:12 AM
19 points
15 comments1 min readLW link

An at­tempt at a “good enough” solu­tion for hu­man two-party negotiations

Isaac KingSep 16, 2023, 12:38 AM
11 points
22 comments1 min readLW link

Nav­i­gat­ing an ecosys­tem that might or might not be bad for the world

Sep 15, 2023, 11:58 PM
79 points
20 comments1 min readLW link

A con­ver­sa­tion with Pi, a con­ver­sa­tional AI.

Spiritus DeiSep 15, 2023, 11:13 PM
1 point
0 comments1 min readLW link

Clos­ing Notes on Non­lin­ear Investigation

Ben PaceSep 15, 2023, 10:44 PM
97 points
47 comments11 min readLW link

Ann Ar­bor, Michi­gan, USA – ACX Mee­tups Every­where Fall 2023

J03MANSep 15, 2023, 9:13 PM
1 point
0 comments1 min readLW link

In­fluence func­tions—why, what and how

Nina PanicksserySep 15, 2023, 8:42 PM
73 points
6 comments8 min readLW link

Ethics Needs A Marginal Revolution

Bentham's BulldogSep 15, 2023, 7:08 PM
23 points
3 comments9 min readLW link

I com­piled a ebook of `Pro­ject Lawful` for eBook readers

OrwellGoesShoppingSep 15, 2023, 6:09 PM
90 points
4 comments1 min readLW link
(www.mikescher.com)

Thoughts on the Waluigi Effect

fibonacchoSep 15, 2023, 5:40 PM
10 points
0 comments12 min readLW link

De­con­fus­ing Regret

Alex HollowSep 15, 2023, 11:52 AM
41 points
32 comments2 min readLW link

From game the­ory to play­ers theory

Victor PortonSep 15, 2023, 6:23 AM
−4 points
0 comments3 min readLW link

SPAR seeks ad­vi­sors and stu­dents for AI safety pro­jects (Se­cond Wave)

micSep 14, 2023, 11:09 PM
21 points
0 comments1 min readLW link

“Did you lock it?”

ymeskhoutSep 14, 2023, 9:10 PM
33 points
36 comments2 min readLW link
(ymeskhout.substack.com)

Can I take ducks home from the park?

dynomightSep 14, 2023, 9:03 PM
67 points
8 comments3 min readLW link
(dynomight.net)

In­line Plot­ting in iTerm2

jefftkSep 14, 2023, 8:30 PM
13 points
0 comments1 min readLW link
(www.jefftk.com)

De­stroy­ing the fabric of the uni­verse as an in­stru­men­tal goal.

AI-doomSep 14, 2023, 8:04 PM
−7 points
5 comments1 min readLW link

The PUSA Sys­tem- Repost

Jaivardhan NawaniSep 14, 2023, 6:40 PM
4 points
1 comment5 min readLW link

# **An­nounce­ment of AI-Plans.com Cri­tique-a-thon Septem­ber 2023**

Kabir KumarSep 14, 2023, 5:43 PM
3 points
0 comments2 min readLW link

Cruxes for overhang

Zach Stein-PerlmanSep 14, 2023, 5:00 PM
12 points
5 comments6 min readLW link
(blog.aiimpacts.org)

A The­ory of Laugh­ter—Fol­low-Up

Steven ByrnesSep 14, 2023, 3:35 PM
37 points
3 comments8 min readLW link

Elic­it­ing Credit Hack­ing Be­havi­ours in LLMs

omegastickSep 14, 2023, 3:07 PM
3 points
2 comments7 min readLW link
(github.com)

In­stru­men­tal Con­ver­gence Bounty

Logan ZoellnerSep 14, 2023, 2:02 PM
62 points
24 comments1 min readLW link

[Question] In the age of mod­ern AI (LLMs and be­yond), is data still the new oil?

MPSep 14, 2023, 1:28 PM
4 points
1 comment2 min readLW link

AI #29: Take a Deep Breath

ZviSep 14, 2023, 12:00 PM
65 points
21 comments21 min readLW link
(thezvi.wordpress.com)

The om­ni­zoid—Heighn FDT De­bate #3: Con­tra om­ni­zoid con­tra me con­tra om­ni­zoid con­tra FDT

HeighnSep 14, 2023, 11:52 AM
6 points
0 comments4 min readLW link

High­lights: Went­worth, Shah, and Mur­phy on “Re­tar­get­ing the Search”

RobertMSep 14, 2023, 2:18 AM
87 points
4 comments8 min readLW link

Un­cov­er­ing La­tent Hu­man Wel­lbe­ing in LLM Embeddings

Sep 14, 2023, 1:40 AM
32 points
7 comments8 min readLW link
(far.ai)

A Call For Com­mu­nity: Scien­tific Lan­guage Learn­ing is Still Lan­guage Learn­ing

keltanSep 14, 2023, 12:32 AM
0 points
0 comments2 min readLW link

Mech In­terp Challenge: Septem­ber—De­ci­pher­ing the Ad­di­tion Model

CallumMcDougallSep 13, 2023, 10:23 PM
35 points
0 comments4 min readLW link

Linkpost for Jan Leike on Self-Exfiltration

Daniel KokotajloSep 13, 2023, 9:23 PM
59 points
1 comment2 min readLW link
(aligned.substack.com)

MLSN: #10 Ad­ver­sar­ial At­tacks Against Lan­guage and Vi­sion Models, Im­prov­ing LLM Hon­esty, and Trac­ing the In­fluence of LLM Train­ing Data

Sep 13, 2023, 6:03 PM
15 points
1 comment5 min readLW link
(newsletter.mlsafety.org)

Ex­pand­ing the Scope of Superposition

Derek LarsonSep 13, 2023, 5:38 PM
10 points
0 comments4 min readLW link

Con­tra Yud­kowsky on Epistemic Con­duct for Author Criticism

Zack_M_DavisSep 13, 2023, 3:33 PM
70 points
38 comments7 min readLW link

Ap­ply to lead a pro­ject dur­ing the next vir­tual AI Safety Camp

Sep 13, 2023, 1:29 PM
19 points
0 comments5 min readLW link
(aisafety.camp)

Is AI Safety drop­ping the ball on pri­vacy?

markovSep 13, 2023, 1:07 PM
50 points
17 comments7 min readLW link

UDT shows that de­ci­sion the­ory is more puz­zling than ever

Wei DaiSep 13, 2023, 12:26 PM
219 points
56 comments1 min readLW link

[Question] Align­ment & Ca­pa­bil­ities: what’s the differ­ence?

johnhalsteadSep 13, 2023, 11:48 AM
6 points
3 comments1 min readLW link

Duty to res­cue /​ Non-as­sis­tance à per­sonne en danger

Thomas SepulchreSep 13, 2023, 9:49 AM
15 points
5 comments3 min readLW link

The Flow-Through Fallacy

Chris_LeongSep 13, 2023, 4:28 AM
21 points
7 comments1 min readLW link

Book re­view: The Im­por­tance of What We Care About (Harry G. Frank­furt)

David GrossSep 13, 2023, 4:17 AM
7 points
0 comments4 min readLW link

Padding the Corner

jefftkSep 13, 2023, 1:30 AM
32 points
4 comments1 min readLW link
(www.jefftk.com)