Look­ing Deeper at Deconfusion

adamShimi13 Jun 2021 21:29 UTC
61 points
13 comments15 min readLW link

[Question] What other prob­lems would a suc­cess­ful AI safety al­gorithm solve?

DirectedEvolution13 Jun 2021 21:07 UTC
12 points
4 comments1 min readLW link

Avoid­ing the in­stru­men­tal policy by hid­ing in­for­ma­tion about humans

paulfchristiano13 Jun 2021 20:00 UTC
31 points
2 comments2 min readLW link

An­swer­ing ques­tions hon­estly given world-model mismatches

paulfchristiano13 Jun 2021 18:00 UTC
34 points
2 comments16 min readLW link
(ai-alignment.com)

Ex­per­i­ments with a ran­dom clock

Malmesbury13 Jun 2021 17:08 UTC
58 points
12 comments7 min readLW link
(www.telescopic-turnip.net)

The Cult Deficit: Anal­y­sis and Speculation

rogersbacon13 Jun 2021 5:33 UTC
18 points
11 comments13 min readLW link

Covid vac­cine safety: how cor­rect are these alle­ga­tions?

DPiepgrass13 Jun 2021 3:08 UTC
45 points
93 comments19 min readLW link

The Moon is Down; I have not heard the clock

chemslug13 Jun 2021 1:40 UTC
52 points
5 comments2 min readLW link

A sys­tem­atic er­ror that lead to a bad policy re­sponse to COVID-19

ChristianKl12 Jun 2021 19:56 UTC
23 points
6 comments3 min readLW link

[Question] Why do pa­tients in men­tal in­sti­tu­tions get so lit­tle at­ten­tion in the pub­lic dis­course?

ChristianKl12 Jun 2021 17:37 UTC
41 points
29 comments1 min readLW link

In­ter­nal Memo from Bleggs Universal

kithpendragon12 Jun 2021 16:44 UTC
24 points
2 comments3 min readLW link

Some peo­ple just won’t get it

Amir Bolous12 Jun 2021 15:10 UTC
3 points
5 comments2 min readLW link

[Question] Is the length of the Covid-19 in­cu­ba­tion pe­riod likely to be af­fected by whether you are vac­ci­nated?

Josh Jacobson12 Jun 2021 3:27 UTC
9 points
15 comments1 min readLW link

[Question] Does any­one have any Data Sid­e­quests?

abstractapplic11 Jun 2021 23:40 UTC
18 points
2 comments1 min readLW link

Com­ment on the lab leak hypothesis

Jeffrey Ladish11 Jun 2021 22:49 UTC
61 points
14 comments4 min readLW link

Overview of Re­think Pri­ori­ties’ work on risks from nu­clear weapons

MichaelA11 Jun 2021 20:05 UTC
12 points
0 comments3 min readLW link

[Question] Self-study ideas for micro-pro­jects in “ab­stract” sub­jects?

DirectedEvolution11 Jun 2021 19:09 UTC
13 points
15 comments1 min readLW link

[Question] Am I anti-so­cial if I get vac­ci­nated now?

FlorianH11 Jun 2021 15:42 UTC
5 points
14 comments1 min readLW link

Alarms Are Bet­ter Than Chivvying

jefftk11 Jun 2021 15:30 UTC
21 points
1 comment1 min readLW link
(www.jefftk.com)

Oh No My AI (Filk)

Gordon Seidoh Worley11 Jun 2021 15:05 UTC
42 points
7 comments1 min readLW link

May 2021 Gw­ern.net newsletter

gwern11 Jun 2021 14:13 UTC
31 points
0 comments1 min readLW link
(www.gwern.net)

Daniel M. In­gram at the As­tralCodexTen On­line Meetup

JoshuaFox11 Jun 2021 11:35 UTC
7 points
0 comments1 min readLW link

Would you like me to de­bug your math?

Gurkenglas11 Jun 2021 10:54 UTC
58 points
16 comments1 min readLW link

[Question] What are the best ways to im­prove re­siliency?

Nicole Ross11 Jun 2021 3:22 UTC
16 points
5 comments1 min readLW link

[Question] How do you keep track of your own learn­ing?

DirectedEvolution10 Jun 2021 20:25 UTC
8 points
8 comments1 min readLW link

Covid 6/​10: Some­body Else’s Problem

Zvi10 Jun 2021 13:10 UTC
57 points
26 comments24 min readLW link
(thezvi.wordpress.com)

[Question] ML is now au­tomat­ing parts of chip R&D. How big a deal is this?

Daniel Kokotajlo10 Jun 2021 9:51 UTC
45 points
17 comments1 min readLW link

“Ne­c­es­sary Claims”: A tech­nique to struc­ture com­plex decisions

fgrosseholz@yahoo.com10 Jun 2021 5:16 UTC
8 points
0 comments6 min readLW link

Re­ply to Nate Soares on Dolphins

Zack_M_Davis10 Jun 2021 4:53 UTC
46 points
49 comments5 min readLW link

The Ap­pren­tice Experiment

johnswentworth10 Jun 2021 3:29 UTC
150 points
11 comments4 min readLW link

Chang­ing my life in 2021, halfway through

Borasko10 Jun 2021 1:36 UTC
42 points
6 comments7 min readLW link

Finite Fac­tored Sets: Orthog­o­nal­ity and Time

Scott Garrabrant10 Jun 2021 1:22 UTC
32 points
9 comments4 min readLW link

Knowl­edge is not just mu­tual information

Alex Flint10 Jun 2021 1:01 UTC
28 points
6 comments4 min readLW link

A naive al­ign­ment strat­egy and op­ti­mism about generalization

paulfchristiano10 Jun 2021 0:10 UTC
44 points
4 comments3 min readLW link
(ai-alignment.com)

What are some im­por­tant in­sights you would give to a younger ver­sion of your­self?

silentbob9 Jun 2021 20:29 UTC
9 points
5 comments4 min readLW link

Other Con­struc­tions of Gravity

michaelcohen9 Jun 2021 20:06 UTC
4 points
2 comments2 min readLW link

[Question] How do you es­tab­lish a com­fort zone in your stud­ies?

DirectedEvolution9 Jun 2021 17:01 UTC
11 points
7 comments1 min readLW link

Book re­view: “Feel­ing Great” by David Burns

Steven Byrnes9 Jun 2021 13:17 UTC
103 points
12 comments8 min readLW link

Bad names make you open the box

Adam Zerner9 Jun 2021 3:19 UTC
90 points
51 comments10 min readLW link1 review

...and then some­times, for no clear rea­son, they in­nately be­come good.

Andrew Vlahos9 Jun 2021 3:07 UTC
2 points
3 comments1 min readLW link

Bet­ter Whis­tle Synth

jefftk9 Jun 2021 1:50 UTC
9 points
0 comments2 min readLW link
(www.jefftk.com)

“De­ci­sion Trans­former” (Tool AIs are se­cret Agent AIs)

gwern9 Jun 2021 1:06 UTC
37 points
4 comments1 min readLW link
(sites.google.com)

AXRP Epi­sode 8 - As­sis­tance Games with Dy­lan Had­field-Menell

DanielFilan8 Jun 2021 23:20 UTC
22 points
1 comment72 min readLW link

Should we vac­ci­nate against PGBD5 which codes for a trans­posase?

ChristianKl8 Jun 2021 19:22 UTC
18 points
5 comments1 min readLW link

Evan Hub­inger on Ho­mo­gene­ity in Take­off Speeds, Learned Op­ti­miza­tion and Interpretability

Michaël Trazzi8 Jun 2021 19:20 UTC
28 points
0 comments55 min readLW link

Sur­vey on AI ex­is­ten­tial risk scenarios

8 Jun 2021 17:12 UTC
63 points
11 comments7 min readLW link

[Question] What are the gears of gluten sen­si­tivity?

Gordon Seidoh Worley8 Jun 2021 16:42 UTC
20 points
20 comments1 min readLW link

The Gen­er­al­ized Product Rule

aysajan8 Jun 2021 16:39 UTC
34 points
7 comments5 min readLW link

[Question] Will out­lets like the NYT be cap­tured by Chi­nese in­fluence and if so, when?

ChristianKl8 Jun 2021 16:31 UTC
20 points
3 comments1 min readLW link

The re­verse Good­hart problem

Stuart_Armstrong8 Jun 2021 15:48 UTC
20 points
22 comments1 min readLW link