Notes from “Don’t Shoot the Dog”

juliawise2 Apr 2021 16:34 UTC
252 points
13 comments12 min readLW link1 review

Another (outer) al­ign­ment failure story

paulfchristiano7 Apr 2021 20:12 UTC
244 points
38 comments12 min readLW link1 review

An­nounc­ing the Align­ment Re­search Center

paulfchristiano26 Apr 2021 23:30 UTC
178 points
6 comments1 min readLW link
(ai-alignment.com)

Pre­dic­tive Cod­ing has been Unified with Backpropagation

lsusr2 Apr 2021 21:42 UTC
176 points
51 comments2 min readLW link

Spe­cial­iz­ing in Prob­lems We Don’t Understand

johnswentworth10 Apr 2021 22:40 UTC
174 points
29 comments8 min readLW link1 review

I’m from a par­allel Earth with much higher co­or­di­na­tion: AMA

5 Apr 2021 22:09 UTC
172 points
35 comments61 min readLW link1 review

Test­ing The Nat­u­ral Ab­strac­tion Hy­poth­e­sis: Pro­ject Intro

johnswentworth6 Apr 2021 21:24 UTC
168 points
41 comments6 min readLW link1 review

Why has nu­clear power been a flop?

jasoncrawford16 Apr 2021 16:49 UTC
148 points
50 comments15 min readLW link2 reviews
(rootsofprogress.org)

The Case for Ex­treme Vac­cine Effectiveness

Ruby13 Apr 2021 21:08 UTC
142 points
37 comments23 min readLW link

Opinions on In­ter­pretable Ma­chine Learn­ing and 70 Sum­maries of Re­cent Papers

9 Apr 2021 19:19 UTC
141 points
17 comments102 min readLW link

“AI and Com­pute” trend isn’t pre­dic­tive of what is happening

alexlyzhov2 Apr 2021 0:44 UTC
133 points
16 comments1 min readLW link

Why We Launched LessWrong.SubStack

Ben Pace1 Apr 2021 6:34 UTC
132 points
44 comments4 min readLW link

Tales from Pre­dic­tion Markets

ike3 Apr 2021 23:38 UTC
128 points
15 comments3 min readLW link1 review
(misinfounderload.substack.com)

AMA: Paul Chris­ti­ano, al­ign­ment researcher

paulfchristiano28 Apr 2021 18:55 UTC
117 points
197 comments1 min readLW link

A new acausal trad­ing plat­form: RobinShould

Matthew Barnett1 Apr 2021 16:56 UTC
116 points
5 comments1 min readLW link

Monastery and Throne

Jacob Falkovich6 Apr 2021 19:00 UTC
115 points
42 comments10 min readLW link

Jaan Tal­linn’s 2020 Philan­thropy Overview

jaan27 Apr 2021 16:22 UTC
113 points
4 comments1 min readLW link
(jaan.online)

The ir­rele­vance of test scores is greatly exaggerated

dynomight15 Apr 2021 14:15 UTC
111 points
13 comments1 min readLW link
(dynomight.net)

How to Play a Sup­port Role in Re­search Conversations

johnswentworth23 Apr 2021 20:57 UTC
105 points
4 comments5 min readLW link

Covid 4/​22: Cri­sis in India

Zvi22 Apr 2021 13:40 UTC
100 points
25 comments12 min readLW link
(thezvi.wordpress.com)

[Let­ter] Ad­vice for High School #1

lsusr20 Apr 2021 4:09 UTC
93 points
28 comments4 min readLW link

High­lights from The Au­to­bi­og­ra­phy of An­drew Carnegie

jasoncrawford8 Apr 2021 22:03 UTC
92 points
9 comments19 min readLW link1 review
(rootsofprogress.org)

“Tak­ing your en­vi­ron­ment as ob­ject” vs “Be­ing sub­ject to your en­vi­ron­ment”

Ben Pace11 Apr 2021 22:47 UTC
86 points
17 comments3 min readLW link

Draft re­port on ex­is­ten­tial risk from power-seek­ing AI

Joe Carlsmith28 Apr 2021 21:41 UTC
85 points
23 comments1 min readLW link

Peo­ple Will Listen

sapphire11 Apr 2021 16:51 UTC
85 points
36 comments4 min readLW link

What are all these chil­dren do­ing in my ponds?

dominicq3 Apr 2021 20:16 UTC
83 points
15 comments3 min readLW link

Covid 4/​15: Are We Se­ri­ously Do­ing This Again

Zvi15 Apr 2021 13:00 UTC
82 points
37 comments9 min readLW link
(thezvi.wordpress.com)

A Brief Re­view of Cur­rent and Near-Fu­ture Meth­ods of Ge­netic Engineering

GeneSmith10 Apr 2021 19:16 UTC
81 points
33 comments15 min readLW link

Gra­da­tions of In­ner Align­ment Obstacles

abramdemski20 Apr 2021 22:18 UTC
81 points
22 comments9 min readLW link

Iter­ated Trust Kickstarters

Raemon20 Apr 2021 3:18 UTC
74 points
19 comments10 min readLW link

Up­dat­ing the Lot­tery Ticket Hypothesis

johnswentworth18 Apr 2021 21:45 UTC
73 points
41 comments2 min readLW link

Want­ing to Suc­ceed on Every Met­ric Presented

Logan Riggs12 Apr 2021 20:43 UTC
72 points
25 comments3 min readLW link

Cen­ter for Ap­plied Pos­tra­tional­ity: An Update

Pee Doom1 Apr 2021 8:13 UTC
71 points
1 comment3 min readLW link

Beliefs as emo­tional strategies

Kaj_Sotala9 Apr 2021 14:28 UTC
70 points
4 comments8 min readLW link

Hell is wasted on the evil

lsusr15 Apr 2021 8:52 UTC
70 points
17 comments3 min readLW link

The se­cret of Wikipe­dia’s success

Aaron Bergman14 Apr 2021 22:18 UTC
68 points
11 comments6 min readLW link
(aaronbergman.substack.com)

FAQ: Ad­vice for AI Align­ment Researchers

Rohin Shah26 Apr 2021 18:59 UTC
67 points
2 comments1 min readLW link
(rohinshah.com)

Agents Over Carte­sian World Models

27 Apr 2021 2:06 UTC
66 points
4 comments27 min readLW link

Don’t Sell Your Soul

Jacob Falkovich6 Apr 2021 19:02 UTC
65 points
43 comments9 min readLW link

Solv­ing the whole AGI con­trol prob­lem, ver­sion 0.0001

Steven Byrnes8 Apr 2021 15:14 UTC
63 points
7 comments26 min readLW link

Reflec­tive Bayesianism

abramdemski6 Apr 2021 19:48 UTC
62 points
27 comments13 min readLW link

A New Cen­ter? [Poli­tics] [Wish­ful Think­ing]

abramdemski12 Apr 2021 15:19 UTC
62 points
36 comments3 min readLW link

Against “Con­text-Free In­tegrity”

Ben Pace14 Apr 2021 8:20 UTC
62 points
28 comments5 min readLW link

My take on Michael Littman on “The HCI of HAI”

Alex Flint2 Apr 2021 19:51 UTC
59 points
4 comments7 min readLW link

Thiel on se­crets and indefiniteness

Rob Bensinger20 Apr 2021 21:59 UTC
58 points
7 comments9 min readLW link

Affordances

abramdemski2 Apr 2021 20:53 UTC
58 points
18 comments6 min readLW link

You Can Now Embed Flash­card Quizzes in Your LessWrong posts!

spencerg19 Apr 2021 13:44 UTC
57 points
25 comments4 min readLW link

Ra­tion­al­ity Cardinality

jimrandomh27 Apr 2021 22:27 UTC
57 points
7 comments1 min readLW link

Covid 4/​29: Vac­ci­na­tion Slowdown

Zvi29 Apr 2021 13:50 UTC
55 points
19 comments15 min readLW link
(thezvi.wordpress.com)

Align­ment Newslet­ter Three Year Retrospective

Rohin Shah7 Apr 2021 14:39 UTC
55 points
0 comments5 min readLW link