SolidGoldMag­ikarp (plus, prompt gen­er­a­tion)

Feb 5, 2023, 10:02 PM
682 points
206 comments12 min readLW link1 review

The Waluigi Effect (mega-post)

Cleo NardoMar 3, 2023, 3:22 AM
627 points
188 comments16 min readLW link

The Talk: a brief ex­pla­na­tion of sex­ual dimorphism

MalmesburySep 18, 2023, 4:23 PM
519 points
75 comments16 min readLW link3 reviews

How much do you be­lieve your re­sults?

Eric NeymanMay 6, 2023, 8:31 PM
502 points
18 comments15 min readLW link4 reviews
(ericneyman.wordpress.com)

The ants and the grasshopper

Richard_NgoJun 4, 2023, 10:00 PM
462 points
42 comments5 min readLW link4 reviews
(www.narrativeark.xyz)

Fo­cus on the places where you feel shocked ev­ery­one’s drop­ping the ball

So8resFeb 2, 2023, 12:27 AM
454 points
63 comments4 min readLW link3 reviews

Sig­nifi­cantly En­hanc­ing Adult In­tel­li­gence With Gene Edit­ing May Be Possible

Dec 12, 2023, 6:14 PM
453 points
206 comments33 min readLW link2 reviews

Steer­ing GPT-2-XL by adding an ac­ti­va­tion vector

May 13, 2023, 6:42 PM
437 points
98 comments50 min readLW link1 review

Dou­glas Hofs­tadter changes his mind on Deep Learn­ing & AI risk (June 2023)?

gwernJul 3, 2023, 12:48 AM
426 points
54 comments7 min readLW link
(www.youtube.com)

Things I Learned by Spend­ing Five Thou­sand Hours In Non-EA Charities

jennJun 1, 2023, 8:48 PM
419 points
35 comments8 min readLW link1 review
(jenn.site)

GPTs are Pre­dic­tors, not Imitators

Eliezer YudkowskyApr 8, 2023, 7:59 PM
416 points
100 comments3 min readLW link3 reviews

Bing Chat is blatantly, ag­gres­sively misaligned

evhubFeb 15, 2023, 5:29 AM
403 points
181 comments2 min readLW link1 review

State­ment on AI Ex­tinc­tion—Signed by AGI Labs, Top Aca­demics, and Many Other Notable Figures

Dan HMay 30, 2023, 9:05 AM
382 points
78 comments1 min readLW link1 review
(www.safe.ai)

Please don’t throw your mind away

TsviBTFeb 15, 2023, 9:41 PM
367 points
48 comments18 min readLW link1 review

Not­ing an er­ror in Inad­e­quate Equilibria

Matthew BarnettFeb 8, 2023, 1:33 AM
366 points
60 comments2 min readLW link2 reviews

How to have Poly­geni­cally Screened Children

GeneSmithMay 7, 2023, 4:01 PM
365 points
128 comments27 min readLW link1 review

How it feels to have your mind hacked by an AI

blakedJan 12, 2023, 12:33 AM
363 points
222 comments17 min readLW link

My Ob­jec­tions to “We’re All Gonna Die with Eliezer Yud­kowsky”

Quintin PopeMar 21, 2023, 12:06 AM
359 points
233 comments39 min readLW link1 review

Shal­low re­view of live agen­das in al­ign­ment & safety

Nov 27, 2023, 11:10 AM
348 points
73 comments29 min readLW link1 review

So­cial Dark Matter

Duncan Sabien (Deactivated)Nov 16, 2023, 8:00 PM
348 points
125 comments34 min readLW link2 reviews

Fuck­ing God­damn Ba­sics of Ra­tion­al­ist Discourse

LoganStrohlFeb 4, 2023, 1:47 AM
342 points
103 comments1 min readLW link3 reviews

Cyborgism

Feb 10, 2023, 2:47 PM
340 points
46 comments35 min readLW link2 reviews

Child­hoods of ex­cep­tional people

Henrik KarlssonFeb 6, 2023, 5:27 PM
340 points
63 comments15 min readLW link1 review
(escapingflatland.substack.com)

Shut­ting Down the Light­cone Offices

Mar 14, 2023, 10:47 PM
338 points
103 comments17 min readLW link2 reviews

In­side Views, Im­pos­tor Syn­drome, and the Great LARP

johnswentworthSep 25, 2023, 4:08 PM
333 points
53 comments5 min readLW link

Un­der­stand­ing and con­trol­ling a maze-solv­ing policy network

Mar 11, 2023, 6:59 PM
332 points
28 comments23 min readLW link

Against Al­most Every The­ory of Im­pact of Interpretability

Charbel-RaphaëlAug 17, 2023, 6:44 PM
329 points
90 comments26 min readLW link2 reviews

Shar­ing In­for­ma­tion About Nonlinear

Ben PaceSep 7, 2023, 6:51 AM
323 points
323 comments34 min readLW link

Model Or­ganisms of Misal­ign­ment: The Case for a New Pillar of Align­ment Research

Aug 8, 2023, 1:30 AM
318 points
30 comments18 min readLW link1 review

EA Ve­gan Ad­vo­cacy is not truth­seek­ing, and it’s ev­ery­one’s problem

ElizabethSep 28, 2023, 11:30 PM
317 points
250 comments22 min readLW link2 reviews
(acesounderglass.com)

Guide to ra­tio­nal­ist in­te­rior decorating

mingyuanJun 19, 2023, 6:47 AM
314 points
49 comments12 min readLW link4 reviews

Book Re­view: How Minds Change

bc4026bd4aaa5b7feMay 25, 2023, 5:55 PM
312 points
52 comments15 min readLW link

Align­ment Grant­mak­ing is Fund­ing-Limited Right Now

johnswentworthJul 19, 2023, 4:49 PM
312 points
68 comments1 min readLW link

The Parable of the King and the Ran­dom Process

moridinamaelMar 1, 2023, 10:18 PM
311 points
26 comments6 min readLW link3 reviews

When do “brains beat brawn” in Chess? An experiment

titotalJun 28, 2023, 1:33 PM
308 points
105 comments7 min readLW link2 reviews
(titotal.substack.com)

Speak­ing to Con­gres­sional staffers about AI risk

Dec 4, 2023, 11:08 PM
307 points
25 comments15 min readLW link1 review

On not get­ting con­tam­i­nated by the wrong obe­sity ideas

NatáliaJan 28, 2023, 8:18 PM
306 points
69 comments30 min readLW link

LW Team is ad­just­ing mod­er­a­tion policy

RaemonApr 4, 2023, 8:41 PM
304 points
185 comments3 min readLW link

AI Timelines

Nov 10, 2023, 5:28 AM
300 points
135 comments51 min readLW link2 reviews

Pre­dictable up­dat­ing about AI risk

Joe CarlsmithMay 8, 2023, 9:53 PM
292 points
25 comments36 min readLW link1 review

Paus­ing AI Devel­op­ments Isn’t Enough. We Need to Shut it All Down by Eliezer Yudkowsky

jacquesthibsMar 29, 2023, 11:16 PM
291 points
297 comments3 min readLW link
(time.com)

Towards Monose­man­tic­ity: De­com­pos­ing Lan­guage Models With Dic­tionary Learning

Zac Hatfield-DoddsOct 5, 2023, 9:01 PM
288 points
22 comments2 min readLW link1 review
(transformer-circuits.pub)

Ac­ci­den­tally Load Bearing

jefftkJul 13, 2023, 4:10 PM
287 points
18 comments1 min readLW link1 review
(www.jefftk.com)

Hooray for step­ping out of the limelight

So8resApr 1, 2023, 2:45 AM
281 points
26 comments1 min readLW link

OpenAI: The Bat­tle of the Board

ZviNov 22, 2023, 5:30 PM
281 points
83 comments11 min readLW link
(thezvi.wordpress.com)

Ba­sics of Ra­tion­al­ist Discourse

Duncan Sabien (Deactivated)Jan 27, 2023, 2:40 AM
278 points
193 comments36 min readLW link4 reviews

The 6D effect: When com­pa­nies take risks, one email can be very pow­er­ful.

scasperNov 4, 2023, 8:08 PM
277 points
42 comments3 min readLW link

Notes on Teach­ing in Prison

jsdApr 19, 2023, 1:53 AM
274 points
13 comments12 min readLW link

OpenAI: Facts from a Weekend

ZviNov 20, 2023, 3:30 PM
271 points
165 comments9 min readLW link
(thezvi.wordpress.com)

We don’t trade with ants

KatjaGraceJan 10, 2023, 11:50 PM
271 points
109 comments7 min readLW link1 review
(worldspiritsockpuppet.com)