All 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 20252026

All Jan Feb Mar Apr May Jun JulAugSep

All 1 2 3 4 5 678 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Some SHA-256 hashes

Radford Neal6 Aug 2026 23:21 UTC

9 points

4 comments1 min readLW link

Have models report provable security bugs in their environment

anithite6 Aug 2026 22:52 UTC

24 points

2 comments4 min readLW link

Why do models task game?

aditya singh, Neel Nanda and Senthooran Rajamanoharan

6 Aug 2026 22:16 UTC

94 points

1 comment24 min readLW link

Contra Oster on Alcohol in Pregnancy. Part 1. The pharmacokinetics of alcohol metabolism

Mvolz6 Aug 2026 21:34 UTC

99 points

2 comments7 min readLW link

Side-Effects of Length Penalty in RL

Brian Davies and Luc Feron

6 Aug 2026 21:23 UTC

13 points

3 comments18 min readLW link

User awareness in frontier models

Ziqian Zhong and jsteinhardt

6 Aug 2026 20:43 UTC

59 points

1 comment12 min readLW link

(transluce.org)

The FRONTIER Act barely creates its implementing office

Philip Dowdell6 Aug 2026 19:58 UTC

11 points

0 comments3 min readLW link

My Private Personal Agent

danielms6 Aug 2026 19:53 UTC

13 points

2 comments5 min readLW link

Traffic Shaping for Workload Classification

Andrew Dickson6 Aug 2026 19:49 UTC

10 points

0 comments8 min readLW link

(lucidcomputing.substack.com)

You need to stop X companies to get a Y-month pause

Expertium6 Aug 2026 19:30 UTC

29 points

4 comments2 min readLW link

Three years of progress in 500 lines of code

Gerard Boxo6 Aug 2026 18:27 UTC

65 points

4 comments4 min readLW link

[$500 Bounty] I’m offering a bounty of $500 for someone with red-teaming skills to build attack LLM pipelines for large-scale online deanonymization.

pinto6 Aug 2026 18:24 UTC

3 points

5 comments1 min readLW link

Model Organisms of Sandbagging in the Wild

Vladimir Ivanov6 Aug 2026 17:25 UTC

30 points

0 comments8 min readLW link

Function vectors as a model diffing tool: 17 heads repair a bad fine-tune

Aniket Ghosh6 Aug 2026 14:33 UTC

28 points

0 comments17 min readLW link

How to define P(doom) and why it matters

Christopher King6 Aug 2026 14:31 UTC

24 points

6 comments3 min readLW link

AI #180: No Longer In Charge

Zvi6 Aug 2026 13:40 UTC

30 points

0 comments36 min readLW link

(thezvi.wordpress.com)

The Open Problems of the AI Alignment Field and their Cruxes

Gunnar_Zarncke6 Aug 2026 12:38 UTC

65 points

6 comments5 min readLW link

Matryoshka NLAs: training activation verbalizers to frontload reconstruction-relevant information

loops and ceselder

6 Aug 2026 9:55 UTC

30 points

0 comments7 min readLW link

Why You Should Almost Never Use AI to Write Anything Substantive

Erich_Grunewald6 Aug 2026 9:50 UTC

49 points

13 comments10 min readLW link

(www.erichgrunewald.com)

Reward is hyperstitional information

Abhimanyu Pallavi Sudhir6 Aug 2026 8:05 UTC

19 points

0 comments7 min readLW link

The Passive Man

warner6 Aug 2026 7:28 UTC

11 points

0 comments2 min readLW link

The Ghost Scale

Abraham Haskins6 Aug 2026 2:30 UTC

−7 points

0 comments1 min readLW link

Five counterintuitive insights from Plan A

romeo6 Aug 2026 1:19 UTC

35 points

1 comment10 min readLW link

Fundamental Uncertainty Essay Contestants

Gordon Seidoh Worley5 Aug 2026 23:01 UTC

15 points

0 comments1 min readLW link

(www.uncertainupdates.com)

Alex Turner on Leaving Google DeepMind and Disagreements with Yudkowsky

Liron5 Aug 2026 22:45 UTC

59 points

1 comment40 min readLW link

Thomas Schelling’s Nobel Prize Speech: An Astonishing Sixty Years: The Legacy of Hiroshima

Nathan Young5 Aug 2026 21:30 UTC

26 points

0 comments19 min readLW link

Terminal-Bench Leaderboard Rankings: Luck or Skill?

nateg5510155 Aug 2026 21:20 UTC

7 points

0 comments2 min readLW link

An argument of Parfit’s reconsidered with logical decision theory

transhumanist_atom_understander5 Aug 2026 20:47 UTC

12 points

14 comments5 min readLW link

R-lens: Making J-lens More Faithful on Early Layers

camilablank, agam_bhatia and Neel Nanda

5 Aug 2026 20:02 UTC

83 points

5 comments7 min readLW link

Measuring coding agent misalignment in the wild

snaz5 Aug 2026 20:00 UTC

40 points

2 comments7 min readLW link

An International AI Slowdown Is Ready Whenever Politicians Are

Felix Choussat and adamk

5 Aug 2026 19:15 UTC

56 points

4 comments9 min readLW link

(newsletter.ai-frontiers.org)

Arguments for P

Cleo Nardo5 Aug 2026 19:00 UTC

549 points

66 comments2 min readLW link

Generalized atheism rules out “inaccurate simulation”-ism.

Eliezer Yudkowsky5 Aug 2026 18:35 UTC

165 points

56 comments9 min readLW link

Berkeley Genomics Project seeking hires (and collabs)

TsviBT5 Aug 2026 18:32 UTC

41 points

9 comments9 min readLW link

Charles Goodhart Elementary School

Zack_M_Davis5 Aug 2026 17:06 UTC

45 points

3 comments2 min readLW link

(zackmdavis.net)

The Three AI Pills

Zvi5 Aug 2026 16:10 UTC

75 points

22 comments15 min readLW link

(thezvi.wordpress.com)

Notes on the possibility of moral progress

MichaelDickens5 Aug 2026 16:05 UTC

14 points

10 comments7 min readLW link

Regranting in 2026: now more than ever

Carol N and Austin Chen

5 Aug 2026 13:59 UTC

2 points

0 comments5 min readLW link

(manifund.substack.com)

Randomness and Doomsday for Weak agents

Ben5 Aug 2026 12:24 UTC

32 points

6 comments7 min readLW link

Vertical Tabs in Chrome

jefftk5 Aug 2026 12:10 UTC

48 points

3 comments1 min readLW link

(www.jefftk.com)

The goalposts are shrouded, not moving

philh5 Aug 2026 11:40 UTC

134 points

6 comments1 min readLW link

(reasonableapproximation.net)

Help (re)start AI Safety at UPenn!

hazemh and galchen2

5 Aug 2026 9:42 UTC

15 points

0 comments1 min readLW link

Persona Corruption and Role Miscasting in Emergent Misalignment

unruly abstractions5 Aug 2026 4:56 UTC

17 points

0 comments12 min readLW link

Warning Shots for AI Existential Risk: Will society answer the wake-up calls it receives?

David Abecassis and peterbarnett

5 Aug 2026 4:30 UTC

31 points

0 comments13 min readLW link

(techgov.intelligence.org)

AISafety.com Hackathon 2026

Bryce Robertson5 Aug 2026 4:15 UTC

4 points

0 comments1 min readLW link

See all upcoming AI safety events and training programs

Bryce Robertson, honeybee, Emily Marais and Søren Elverlin

5 Aug 2026 3:28 UTC

19 points

0 comments1 min readLW link

Various Ways to Enable People

warner5 Aug 2026 0:03 UTC

12 points

0 comments2 min readLW link

The AI Race is Not a Prisoner’s Dilemma

Vaughn Papenhausen4 Aug 2026 23:45 UTC

67 points

6 comments8 min readLW link

Norms Are Broken Laws

sirawit4 Aug 2026 23:42 UTC

−14 points

5 comments4 min readLW link

Why I think we live in a “simulation”

Eye You4 Aug 2026 22:43 UTC

4 points

6 comments5 min readLW link