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

All JanFebMar Apr May Jun Jul Aug Sep

All 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 262728

Anthropic: “Statement from Dario Amodei on our discussions with the Department of War”

Matrice Jacobine26 Feb 2026 23:45 UTC

159 points

22 comments3 min readLW link

(www.anthropic.com)

Asymmetric Risks of Unfaithful Reasoning: Omission as the Critical Failure Mode for AI Monitoring

Divyansh Singhvi26 Feb 2026 21:22 UTC

7 points

0 comments4 min readLW link

Getting Back To It

sarahconstantin26 Feb 2026 20:30 UTC

38 points

1 comment7 min readLW link

(sarahconstantin.substack.com)

The Voices That Are Missing From Sex-Themed Online Communities

Bowl of Cereal26 Feb 2026 20:23 UTC

−19 points

6 comments1 min readLW link

Inference-time Generative Debates on Coding and Reasoning Tasks for Scalable Oversight

ethanelasky and f_niv

26 Feb 2026 20:11 UTC

8 points

0 comments6 min readLW link

A minor point about instrumental convergence that I would like feedback on

agrippa26 Feb 2026 19:44 UTC

4 points

5 comments2 min readLW link

AI welfare as a demotivator for takeover.

Valentin202626 Feb 2026 18:31 UTC

5 points

0 comments3 min readLW link

Frontier AI companies probably can’t leave the US

Anders Cairns Woodruff26 Feb 2026 18:18 UTC

137 points

19 comments7 min readLW link

(blog.redwoodresearch.org)

Improving Internal Model Principle

mremre26 Feb 2026 17:33 UTC

15 points

0 comments11 min readLW link

A Positive Case for Faithfulness: LLM Self-Explanations Help Predict Model Behavior

harrymayne, Justin kang, Dewi Gould and noahys

26 Feb 2026 17:03 UTC

29 points

0 comments4 min readLW link

How Robust Is Monitoring Against Secret Loyalties?

Joe Kwon26 Feb 2026 15:50 UTC

8 points

0 comments5 min readLW link

UFO Aliens Are Your Gods

Lord Dreadwar26 Feb 2026 13:32 UTC

−49 points

18 comments4 min readLW link

AI #157: Burn the Boats

Zvi26 Feb 2026 13:30 UTC

48 points

12 comments58 min readLW link

(thezvi.wordpress.com)

How eval awareness might emerge in training

Igor Ivanov26 Feb 2026 10:59 UTC

27 points

12 comments6 min readLW link

Strategic nuclear war twice as likely to occur by accident than by AI decisions according to new study

kromem26 Feb 2026 8:29 UTC

43 points

1 comment5 min readLW link

What is Claude?

epicurus26 Feb 2026 4:26 UTC

14 points

0 comments7 min readLW link

Why is Anthropic is okay with being used for disinformation?

ChristianKl26 Feb 2026 4:20 UTC

13 points

6 comments1 min readLW link

Scoop: Pentagon takes first step toward blacklisting Anthropic

Matrice Jacobine26 Feb 2026 3:10 UTC

15 points

1 comment1 min readLW link

(www.axios.com)

Transformers Have Computational Signatures Orthogonal to Semantic Content

luxia26 Feb 2026 2:55 UTC

10 points

2 comments13 min readLW link

Alignment as Neural Integration: AI as a Cognitive Layer Accountable to Human Limbic Grounding

Ian Williams26 Feb 2026 2:51 UTC

2 points

1 comment7 min readLW link

Investing in light of AI risk

AshL26 Feb 2026 2:51 UTC

7 points

0 comments5 min readLW link

Whack-a-Mole is Not a Winnable Game

Sable26 Feb 2026 2:40 UTC

105 points

26 comments18 min readLW link

(affablyevil.substack.com)

Announcing ControlConf 2026

Buck26 Feb 2026 2:23 UTC

82 points

4 comments2 min readLW link

Ensuring Safety in Mixed Deployment

Cleo Nardo26 Feb 2026 2:15 UTC

22 points

0 comments5 min readLW link

Map the Future Before You Build It

Molly and Deger Turan

26 Feb 2026 1:50 UTC

11 points

0 comments2 min readLW link

(www.metaculus.com)

Schmidt Sciences’ request for proposals on the Science of Trustworthy AI

James Fox25 Feb 2026 21:42 UTC

31 points

0 comments12 min readLW link

(schmidtsciences.smapply.io)

Naloe: A True Program Editor

TristanTrim25 Feb 2026 21:08 UTC

8 points

4 comments3 min readLW link

Anthropic and the Department of War

Zvi25 Feb 2026 21:00 UTC

89 points

10 comments33 min readLW link

(thezvi.wordpress.com)

Does the First Amendment protect Anthropic from Hegseth?

TFD25 Feb 2026 21:00 UTC

10 points

0 comments2 min readLW link

(www.thefloatingdroid.com)

Character Training Induces Motivation Clarification: A Clue to Claude 3 Opus

Oliver Daniels25 Feb 2026 19:43 UTC

82 points

5 comments8 min readLW link

What secret goals does Claude think it has?

loops25 Feb 2026 19:22 UTC

99 points

11 comments4 min readLW link

Splitting the Sun Equally

Commander Zander25 Feb 2026 18:49 UTC

8 points

1 comment3 min readLW link

Reasoning Traces as a Path to Data-Efficient Generalization in Data Poisoning

Joe Kwon25 Feb 2026 18:17 UTC

14 points

0 comments3 min readLW link

Training Agents to Self-Report Misbehavior

Bruce W. Lee, Yueh Han "John" Chen and Tomek Korbak

25 Feb 2026 17:50 UTC

32 points

0 comments8 min readLW link

Why American Politics is Different Now (for Richard Ngo)

Shiva's Right Foot25 Feb 2026 17:42 UTC

1 point

13 comments4 min readLW link

Beyond Moloch: The view from Evolutionary Game Theory

Jonah Wilberg25 Feb 2026 16:25 UTC

23 points

3 comments8 min readLW link

Uncertain Updates: February 2026

Gordon Seidoh Worley25 Feb 2026 16:10 UTC

9 points

2 comments1 min readLW link

(www.uncertainupdates.com)

Praise the Moloch!

Dentosal25 Feb 2026 12:15 UTC

−16 points

2 comments2 min readLW link

Against Epistemic Humility and for Epistemic Precision

PranavG and Gabriel Alfour

25 Feb 2026 11:13 UTC

13 points

1 comment12 min readLW link

(cognition.cafe)

Review: The Cape Town Observatory

spookyuser25 Feb 2026 10:22 UTC

12 points

0 comments8 min readLW link

The Iron Kaleidoscope

edgecase6425 Feb 2026 6:24 UTC

2 points

0 comments2 min readLW link

Prosaic Continual Learning

HunterJay25 Feb 2026 6:11 UTC

44 points

16 comments7 min readLW link

Rumination is a habit (and you can break it!)

Declan Molony25 Feb 2026 2:57 UTC

24 points

5 comments3 min readLW link

In-context learning alone can induce weird generalisation

Cozmin Ududec, Benji Berczi and Kyuhee Kim

25 Feb 2026 2:46 UTC

74 points

3 comments8 min readLW link

On the phenomenological shift known as ‘stream entry’ and its implications for consciousness

cube_flipper25 Feb 2026 1:30 UTC

40 points

6 comments25 min readLW link

(smoothbrains.net)

How to grow a nuke

RomanS25 Feb 2026 0:53 UTC

26 points

1 comment2 min readLW link

A simple rule for causation

Vivek Hebbar24 Feb 2026 23:14 UTC

37 points

2 comments3 min readLW link

SWE-Bench Pro is even worse

Jonathan Gabor24 Feb 2026 22:51 UTC

32 points

2 comments1 min readLW link

(jonathanpgabor.substack.com)

We are all legal realists now

TFD24 Feb 2026 21:51 UTC

−12 points

1 comment4 min readLW link

(www.thefloatingdroid.com)

Responsible Scaling Policy v3

HoldenKarnofsky24 Feb 2026 20:20 UTC

184 points

83 comments36 min readLW link