RSS

Honey is Good

G Wood12 Jun 2026 4:07 UTC
4 points
0 comments3 min readLW link

The Aes­thet­i­cis­ing Vice by Paul Seabright

Linch12 Jun 2026 2:20 UTC
9 points
1 comment2 min readLW link

Ce­lene’s thoughts on consciousness

ToasterLightning12 Jun 2026 0:55 UTC
13 points
9 comments18 min readLW link
(terminuspoint.substack.com)

Con­struct val­idity of Claude Opus 4.8′s Sys­tem Card – A com­men­tary

Maria Federica Martino Lena 11 Jun 2026 23:33 UTC
7 points
0 comments16 min readLW link

you won’t one-shot a perfect sys­tem, but try anyway

PossiblyElaine11 Jun 2026 22:43 UTC
8 points
0 comments4 min readLW link
(possiblyelaine.substack.com)

The long arc of al­ign­ment: sec­ond-or­der in­stru­men­tal con­ver­gence

Emma Leonhart11 Jun 2026 21:12 UTC
0 points
0 comments3 min readLW link

New­comb’s prob­lem from the grand-sys­tem and petty-sys­tem views

transhumanist_atom_understander11 Jun 2026 20:58 UTC
10 points
0 comments5 min readLW link

[New Paper] Pri­ori­tiz­ing Risks from AI: A Delphi Study of 272 Experts

peterslattery11 Jun 2026 20:57 UTC
14 points
0 comments2 min readLW link
(airisk.mit.edu)

Telepa­thy Is (Al­gorith­mi­cally) Easy

Elliot Callender11 Jun 2026 20:31 UTC
5 points
1 comment4 min readLW link

Mort­gage rate: 6.5% If in­dexed: 1.2%. Three No­belists ap­prove.

Bruce Middleton11 Jun 2026 20:31 UTC
4 points
2 comments2 min readLW link

Failing to Rage­bait the New Gemma

11 Jun 2026 17:50 UTC
28 points
0 comments3 min readLW link

Draw­ing Big Bright Lines for Cy­ber & Biolog­i­cal AI

Austin Morrissey11 Jun 2026 0:55 UTC
−5 points
0 comments4 min readLW link

Pre­dic­tive Pro­cess­ing: Con­scious when Training

Chamod Kalupahana11 Jun 2026 0:06 UTC
12 points
1 comment2 min readLW link

Thoughts on Claude Fable’s silent safeguards

Andy Arditi10 Jun 2026 23:35 UTC
47 points
16 comments10 min readLW link

Notes on Algorithms

Menotim10 Jun 2026 23:28 UTC
7 points
0 comments25 min readLW link

An­thropic did not call for a pause on AI

10 Jun 2026 20:02 UTC
72 points
4 comments5 min readLW link
(controlai.news)

Es­ti­mat­ing No-CoT Task-Com­ple­tion Time Hori­zons of Fron­tier AI Models

10 Jun 2026 17:58 UTC
151 points
2 comments4 min readLW link

Th­ese Three Thaumata

chaosmage10 Jun 2026 16:42 UTC
14 points
0 comments1 min readLW link

You Can Catch Sleeper Agents by Teach­ing Another Model to Imi­tate Them

RobinHa10 Jun 2026 15:21 UTC
50 points
2 comments9 min readLW link
(robinhaselhorst.com)

Ma­chinic Psy­chophar­ma­col­ogy: Do LLMs Self-Med­i­cate?

10 Jun 2026 14:15 UTC
116 points
7 comments23 min readLW link