RSS

English writes num­bers backwards

TurnTroutJul 25, 2025, 11:00 PM
10 points
8 comments12 min readLW link
(turntrout.com)

a 9-week trip on retatrutide

AnnaJoJul 25, 2025, 9:41 PM
11 points
0 comments10 min readLW link

[Question] Non-mal­i­cious AI busi­ness ideas open thread /​compilation

lemonhopeJul 25, 2025, 8:31 PM
11 points
2 comments1 min readLW link

[Question] What are the two con­tra­dic­tory the­o­ries of how to eval­u­ate coun­ter­fac­tu­als?

Said AchmizJul 25, 2025, 6:43 PM
25 points
10 comments1 min readLW link

HPMOR: The (Prob­a­bly) Un­told Lore

Jul 25, 2025, 6:39 PM
189 points
34 comments38 min readLW link

Why I Just Took The Giv­ing What We Can Pledge

Bentham's BulldogJul 25, 2025, 4:24 PM
−21 points
0 comments3 min readLW link

Ac­cess to agent CoT makes mon­i­tors vuln­er­a­ble to persuasion

Jul 25, 2025, 4:09 PM
18 points
0 comments4 min readLW link

Con­crete Pro­jects for Im­prov­ing Cur­rent Tech­ni­cal Safety Re­search Automation

Jul 25, 2025, 2:49 PM
9 points
0 comments8 min readLW link

In­tro­duc­ing SB53.info

MKodamaJul 25, 2025, 2:48 PM
5 points
1 comment7 min readLW link

A web­site to cre­ate bets with strangers

biceJul 25, 2025, 11:06 AM
4 points
1 comment1 min readLW link

PTF 102: Con­di­tion­al­iza­tion and Events

Ape in the coatJul 25, 2025, 6:07 AM
8 points
0 comments8 min readLW link

We Built a Tool to Pro­tect Your Dataset From Sim­ple Scrapers

Jul 25, 2025, 5:44 AM
49 points
8 comments3 min readLW link

The Lev­er­age Cycle

AnnapurnaJul 24, 2025, 9:02 PM
13 points
0 comments3 min readLW link
(jorgevelez.substack.com)

Recom­men­da­tions for fu­ture AI growth: from ex­po­nen­tial to lin­ear, with eco­nomic anchors

ZaborJul 24, 2025, 8:11 PM
6 points
0 comments2 min readLW link

Build­ing and eval­u­at­ing al­ign­ment au­dit­ing agents

Jul 24, 2025, 7:22 PM
44 points
1 comment5 min readLW link

Ful­lrank: Bayesian Noisy Sorting

Max NiedermanJul 24, 2025, 7:03 PM
17 points
2 comments3 min readLW link

The Ide­olog­i­cal Spiral

Jul 24, 2025, 1:00 PM
11 points
1 comment10 min readLW link
(cognition.cafe)

Su­per­in­tel­li­gence isn’t Ap­prox­i­mated by a Ra­tional Agent

Nicolas VillarrealJul 24, 2025, 11:41 AM
14 points
4 comments12 min readLW link

Tak­ing Abun­dance Seriously

eeeeeJul 24, 2025, 9:36 AM
33 points
14 comments12 min readLW link

So Shrieked ZAR

balrajsinghJul 23, 2025, 11:25 PM
6 points
0 comments8 min readLW link