(The) Light­cone is noth­ing with­out its peo­ple: LW + Lighthaven’s big fundraiser

habrykaNov 30, 2024, 2:55 AM
609 points
269 comments42 min readLW link

LessWrong’s (first) album: I Have Been A Good Bing

Apr 1, 2024, 7:33 AM
569 points
180 comments11 min readLW link

OpenAI Email Archives (from Musk v. Alt­man and OpenAI blog)

habrykaNov 16, 2024, 6:38 AM
530 points
80 comments51 min readLW link

Align­ment Fak­ing in Large Lan­guage Models

Dec 18, 2024, 5:19 PM
483 points
74 comments10 min readLW link

I would have shit in that alley, too

Declan MolonyJun 18, 2024, 4:41 AM
458 points
135 comments4 min readLW link

The Best Tacit Knowl­edge Videos on Every Subject

Parker ConleyMar 31, 2024, 5:14 PM
420 points
156 comments16 min readLW link

Trans­form­ers Rep­re­sent Belief State Geom­e­try in their Resi­d­ual Stream

Adam ShaiApr 16, 2024, 9:16 PM
416 points
100 comments12 min readLW link

Failures in Kindness

silentbobMar 26, 2024, 9:30 PM
415 points
60 comments9 min readLW link

Reli­able Sources: The Story of David Gerard

TracingWoodgrainsJul 10, 2024, 7:50 PM
390 points
54 comments43 min readLW link

How I got 4.2M YouTube views with­out mak­ing a sin­gle video

Closed Limelike CurvesSep 3, 2024, 3:52 AM
388 points
36 comments1 min readLW link

The hos­tile telepaths problem

ValentineOct 27, 2024, 3:26 PM
383 points
89 comments15 min readLW link

There is way too much serendipity

MalmesburyJan 19, 2024, 7:37 PM
375 points
56 comments7 min readLW link

My hour of mem­o­ryless lucidity

Eric NeymanMay 4, 2024, 1:40 AM
368 points
35 comments5 min readLW link
(ericneyman.wordpress.com)

Sur­vival with­out dignity

L Rudolf LNov 4, 2024, 2:29 AM
360 points
29 comments15 min readLW link
(nosetgauge.substack.com)

Safety isn’t safety with­out a so­cial model (or: dis­pel­ling the myth of per se tech­ni­cal safety)

Andrew_CritchJun 14, 2024, 12:16 AM
357 points
38 comments4 min readLW link

Re­view: Planecrash

L Rudolf LDec 27, 2024, 2:18 PM
357 points
45 comments21 min readLW link
(nosetgauge.substack.com)

No­tifi­ca­tions Re­ceived in 30 Minutes of Class

tanagrabeastMay 26, 2024, 5:02 PM
356 points
16 comments8 min readLW link

Thoughts on seed oil

dynomightApr 20, 2024, 12:29 PM
352 points
129 comments17 min readLW link
(dynomight.net)

[April Fools’ Day] In­tro­duc­ing Open As­teroid Impact

LinchApr 1, 2024, 8:14 AM
336 points
29 comments1 min readLW link
(openasteroidimpact.org)

Biolog­i­cal risk from the mir­ror world

jasoncrawfordDec 12, 2024, 7:07 PM
333 points
37 comments7 min readLW link
(newsletter.rootsofprogress.org)

What Goes Without Saying

sarahconstantinDec 20, 2024, 6:00 PM
330 points
29 comments5 min readLW link
(sarahconstantin.substack.com)

You don’t know how bad most things are nor pre­cisely how they’re bad.

Solenoid_EntityAug 4, 2024, 2:12 PM
325 points
49 comments5 min readLW link

MIRI 2024 Com­mu­ni­ca­tions Strategy

Gretta DulebaMay 29, 2024, 7:33 PM
325 points
216 comments7 min readLW link

I got dysen­tery so you don’t have to

eukaryoteOct 22, 2024, 4:55 AM
320 points
6 comments17 min readLW link
(eukaryotewritesblog.com)

Univer­sal Ba­sic In­come and Poverty

Eliezer YudkowskyJul 26, 2024, 7:23 AM
317 points
137 comments9 min readLW link

Gentle­ness and the ar­tifi­cial Other

Joe CarlsmithJan 2, 2024, 6:21 PM
313 points
33 comments11 min readLW link

Would catch­ing your AIs try­ing to es­cape con­vince AI de­vel­op­ers to slow down or un­de­ploy?

BuckAug 26, 2024, 4:46 PM
308 points
77 comments4 min readLW link

Sleeper Agents: Train­ing De­cep­tive LLMs that Per­sist Through Safety Training

Jan 12, 2024, 7:51 PM
305 points
95 comments3 min readLW link
(arxiv.org)

The Field of AI Align­ment: A Post­mortem, and What To Do About It

johnswentworthDec 26, 2024, 6:48 PM
295 points
160 comments8 min readLW link

Scale Was All We Needed, At First

Gabe MFeb 14, 2024, 1:49 AM
295 points
34 comments8 min readLW link
(aiacumen.substack.com)

Non-Dis­par­age­ment Ca­naries for OpenAI

May 30, 2024, 7:20 PM
288 points
51 comments2 min readLW link

The On­line Sports Gam­bling Ex­per­i­ment Has Failed

ZviNov 11, 2024, 2:30 PM
284 points
59 comments11 min readLW link
(thezvi.wordpress.com)

By de­fault, cap­i­tal will mat­ter more than ever af­ter AGI

L Rudolf LDec 28, 2024, 5:52 PM
282 points
100 comments16 min readLW link
(nosetgauge.substack.com)

My AI Model Delta Com­pared To Yudkowsky

johnswentworthJun 10, 2024, 4:12 PM
280 points
103 comments4 min readLW link

Overview of strong hu­man in­tel­li­gence am­plifi­ca­tion methods

TsviBTOct 8, 2024, 8:37 AM
278 points
144 comments10 min readLW link

Ori­ent­ing to 3 year AGI timelines

Nikola JurkovicDec 22, 2024, 1:15 AM
276 points
51 comments8 min readLW link

The case for en­sur­ing that pow­er­ful AIs are controlled

Jan 24, 2024, 4:11 PM
274 points
71 comments28 min readLW link

Rais­ing chil­dren on the eve of AI

juliawiseFeb 15, 2024, 9:28 PM
274 points
47 comments5 min readLW link

80,000 hours should re­move OpenAI from the Job Board (and similar EA orgs should do similarly)

RaemonJul 3, 2024, 8:34 PM
274 points
71 comments1 min readLW link

“No-one in my org puts money in their pen­sion”

TobesFeb 16, 2024, 6:33 PM
270 points
16 comments9 min readLW link
(seekingtobejolly.substack.com)

On green

Joe CarlsmithMar 21, 2024, 5:38 PM
269 points
35 comments31 min readLW link

Ex­press in­ter­est in an “FHI of the West”

habrykaApr 18, 2024, 3:32 AM
268 points
41 comments3 min readLW link

The Great Data In­te­gra­tion Schlep

sarahconstantinSep 13, 2024, 3:40 PM
268 points
16 comments9 min readLW link
(sarahconstantin.substack.com)

Leav­ing MIRI, Seek­ing Funding

abramdemskiAug 8, 2024, 6:32 PM
264 points
19 comments2 min readLW link

Get­ting 50% (SoTA) on ARC-AGI with GPT-4o

ryan_greenblattJun 17, 2024, 6:44 PM
262 points
50 comments13 min readLW link

My PhD the­sis: Al­gorith­mic Bayesian Epistemology

Eric NeymanMar 16, 2024, 10:56 PM
261 points
14 comments7 min readLW link
(arxiv.org)

Laz­i­ness death spirals

PatrickDFarleySep 19, 2024, 3:58 PM
259 points
40 comments8 min readLW link

Paul Chris­ti­ano named as US AI Safety In­sti­tute Head of AI Safety

Joel BurgetApr 16, 2024, 4:22 PM
256 points
58 comments1 min readLW link
(www.commerce.gov)

the case for CoT un­faith­ful­ness is overstated

nostalgebraistSep 29, 2024, 10:07 PM
253 points
43 comments11 min readLW link

The Best Lay Ar­gu­ment is not a Sim­ple English Yud Essay

J BostockSep 10, 2024, 5:34 PM
252 points
15 comments5 min readLW link