Draft re­port on ex­is­ten­tial risk from power-seek­ing AI

Joe CarlsmithApr 28, 2021, 9:41 PM
85 points
23 comments1 min readLW link

Notes on Robert McIn­tyre’s Brain Preser­va­tion Talk at the Long Now Foundation

JohnGreerApr 28, 2021, 8:47 PM
14 points
2 comments13 min readLW link

AMA: Paul Chris­ti­ano, al­ign­ment researcher

paulfchristianoApr 28, 2021, 6:55 PM
117 points
197 comments1 min readLW link

[Question] Is sit­ting in the sun much bet­ter than sit­ting in the shade?

just_browsingApr 28, 2021, 6:48 PM
16 points
5 comments1 min readLW link

[AN #148]: An­a­lyz­ing gen­er­al­iza­tion across more axes than just ac­cu­racy or loss

Rohin ShahApr 28, 2021, 6:30 PM
24 points
5 comments11 min readLW link
(mailchi.mp)

Face­book is Si­mu­lacra Level 3, An­dreessen is Level 4

Ben PaceApr 28, 2021, 5:38 PM
52 points
12 comments2 min readLW link

When Should the Fire Alarm Go Off: A model for op­ti­mal thresholds

peterbarnettApr 28, 2021, 12:27 PM
40 points
4 comments5 min readLW link
(peterbarnett.org)

Bet­ter air is the eas­iest way not to die

digital_carverApr 28, 2021, 10:24 AM
24 points
9 comments1 min readLW link
(dynomight.net)

Nam­ing and poin­ter thickness

Adam ZernerApr 28, 2021, 6:35 AM
22 points
5 comments2 min readLW link

Talk­ers and doers

Amir BolousApr 28, 2021, 4:26 AM
−1 points
3 comments1 min readLW link

A ca­sual in­tro to Geo­met­ric Algebra

OrfeasApr 28, 2021, 12:00 AM
24 points
2 comments6 min readLW link

Faerie Ring meet #2: At Gather.Town

hamnoxApr 27, 2021, 11:17 PM
15 points
0 comments1 min readLW link

Ra­tion­al­ity Cardinality

jimrandomhApr 27, 2021, 10:27 PM
57 points
7 comments1 min readLW link

Pit­falls of the agent model

Alex FlintApr 27, 2021, 10:19 PM
25 points
5 comments20 min readLW link

[Question] Covid-19 in In­dia: Why didn’t it hap­pen ear­lier?

SherrinfordApr 27, 2021, 7:13 PM
14 points
5 comments1 min readLW link

Jaan Tal­linn’s 2020 Philan­thropy Overview

jaanApr 27, 2021, 4:22 PM
113 points
4 comments1 min readLW link
(jaan.online)

Scott Alexan­der 2021 Pre­dic­tions: Mar­ket Prices

SimonMApr 27, 2021, 2:03 PM
51 points
5 comments2 min readLW link

Modern Mone­tary The­ory for Dummies

theflowerpotApr 27, 2021, 1:11 PM
7 points
5 comments5 min readLW link

[Question] Can you im­prove IQ by prac­tic­ing IQ tests?

ForensicOceanographyApr 27, 2021, 11:28 AM
8 points
43 comments1 min readLW link

Agents Over Carte­sian World Models

Apr 27, 2021, 2:06 AM
67 points
4 comments27 min readLW link

[Question] What top­ics are on Dath Ilan’s civics exam?

jimrandomhApr 27, 2021, 12:59 AM
55 points
38 comments1 min readLW link

Scott Alexan­der 2021 Pre­dic­tions: Buy/​Sell/​Hold

ZviApr 27, 2021, 12:40 AM
42 points
39 comments12 min readLW link
(thezvi.wordpress.com)

An­nounc­ing the Align­ment Re­search Center

paulfchristianoApr 26, 2021, 11:30 PM
178 points
6 comments1 min readLW link
(ai-alignment.com)

[Linkpost] Treach­er­ous turns in the wild

Mark XuApr 26, 2021, 10:51 PM
31 points
6 comments1 min readLW link
(lukemuehlhauser.com)

FAQ: Ad­vice for AI Align­ment Researchers

Rohin ShahApr 26, 2021, 6:59 PM
67 points
2 comments1 min readLW link
(rohinshah.com)

Spoiler-Free Re­views: Mon­ster Slay­ers, Dream Quest and Loop Hero

ZviApr 26, 2021, 3:00 PM
16 points
2 comments3 min readLW link
(thezvi.wordpress.com)

Bayesian and Fre­quen­tists Ap­proaches to Statis­ti­cal Analysis

Christopher ClarkApr 26, 2021, 12:18 AM
4 points
2 comments8 min readLW link

Be­ware over-use of the agent model

Alex FlintApr 25, 2021, 10:19 PM
28 points
10 comments5 min readLW link1 review

“Who I am” is an ax­iom.

dadadarrenApr 25, 2021, 9:59 PM
7 points
8 comments1 min readLW link

Would robots care about Mean­ing and Re­lat­ing?

RaemonApr 25, 2021, 9:17 PM
33 points
24 comments7 min readLW link

[Question] For mRNA vac­cines, is (short-term) effi­cacy re­ally higher af­ter the sec­ond dose?

Sam MarksApr 25, 2021, 8:21 PM
27 points
13 comments3 min readLW link

[Question] Is there a good soft­ware solu­tion for math­e­mat­i­cal ques­tions?

ChristianKlApr 25, 2021, 6:03 PM
9 points
7 comments1 min readLW link

[May – weekly] Fo­cus­mate: iden­tity preservation

Mati_RoyApr 25, 2021, 5:58 PM
6 points
0 comments1 min readLW link

The Fall of Rome, III: Progress Did Not Exist

LukeOnlineApr 25, 2021, 5:54 AM
30 points
18 comments5 min readLW link

[Question] Best em­piri­cal ev­i­dence on bet­ter than SP500 in­vest­ment re­turns?

bluefalconApr 25, 2021, 5:13 AM
12 points
23 comments1 min readLW link

Mal­i­cious non-state ac­tors and AI safety

ketiApr 25, 2021, 3:21 AM
2 points
14 comments2 min readLW link

[Question] Es­ti­mat­ing COVID cases & deaths in In­dia over the com­ing months

GeneSmithApr 24, 2021, 9:35 PM
33 points
8 comments1 min readLW link

Iden­tity in What You Are Not

Gordon Seidoh WorleyApr 24, 2021, 8:11 PM
11 points
1 comment6 min readLW link

Nat­u­ral­ism and AI alignment

Michele CampoloApr 24, 2021, 4:16 PM
5 points
12 comments8 min readLW link

Paper re­view: A Cryp­to­graphic Solu­tion to a Game The­o­retic Problem

victorsintnicolaasApr 24, 2021, 11:54 AM
23 points
4 comments6 min readLW link

Elec­tion re­sults in Cen­tral Europe match some pre-WW1 borders

ForensicOceanographyApr 24, 2021, 6:24 AM
19 points
2 comments3 min readLW link

[Question] How to test tiny skills?

Nicholas KrossApr 24, 2021, 5:02 AM
5 points
6 comments1 min readLW link

[Question] Apps /​ tools for com­bat­ting sleep pro­cras­ti­na­tion?

emily.fanApr 24, 2021, 1:40 AM
3 points
3 comments1 min readLW link

Vac­ci­na­tion and House Rules

jefftkApr 24, 2021, 1:00 AM
33 points
8 comments2 min readLW link
(www.jefftk.com)

Let’s Re­name Our­selves The “Me­tacog­ni­tive Move­ment”

DirectedEvolutionApr 23, 2021, 9:06 PM
52 points
47 comments7 min readLW link

How to Play a Sup­port Role in Re­search Conversations

johnswentworthApr 23, 2021, 8:57 PM
105 points
4 comments5 min readLW link

The Fall of Rome, II: En­ergy Prob­lems?

LukeOnlineApr 23, 2021, 5:58 PM
29 points
8 comments3 min readLW link

The Fall of Rome: Why It’s Rele­vant, And Why We’re Mistaken

LukeOnlineApr 23, 2021, 12:15 PM
40 points
33 comments3 min readLW link

Prob­a­bil­ity the­ory and log­i­cal in­duc­tion as lenses

Alex FlintApr 23, 2021, 2:41 AM
43 points
7 comments6 min readLW link

[Question] Is there any­thing that can stop AGI de­vel­op­ment in the near term?

Wulky WilkinsenApr 22, 2021, 8:37 PM
5 points
5 comments1 min readLW link