All 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 20242025

All Jan Feb Mar Apr May Jun Jul AugSepOct

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 252627 28 29 30

CFAR update, and New CFAR workshops

AnnaSalamon25 Sep 2025 21:12 UTC

197 points

45 comments8 min readLW link

Making Sense of Consciousness Part 5: Consciousness and the Self

sarahconstantin25 Sep 2025 21:10 UTC

13 points

0 comments10 min readLW link

(sarahconstantin.substack.com)

Widening AI Safety’s talent pipeline by meeting people where they are

Ruben Castaing, yanni kyriacos, Nelson Gardner-Challis and danwil

25 Sep 2025 20:50 UTC

30 points

3 comments8 min readLW link

Synthesizing Standalone World-Models, Part 3: Dataset-Assembly

Thane Ruthenis25 Sep 2025 19:21 UTC

13 points

0 comments2 min readLW link

Why you should eat meat—even if you hate factory farming

KatWoods25 Sep 2025 15:39 UTC

299 points

88 comments10 min readLW link

What GPT-oss Leaks About OpenAI’s Training Data

Lennart Finke25 Sep 2025 15:33 UTC

26 points

5 comments6 min readLW link

The real AI deploys itself

David Scott Krueger (formerly: capybaralet)25 Sep 2025 14:11 UTC

76 points

8 comments3 min readLW link

(therealartificialintelligence.substack.com)

AI #135: OpenAI Shows Us The Money

Zvi25 Sep 2025 13:40 UTC

23 points

2 comments44 min readLW link

(thezvi.wordpress.com)

Celebrate Petrov day as if the button had been pressed

Flying buttress25 Sep 2025 10:33 UTC

17 points

0 comments1 min readLW link

Understanding the state of frontier AI in China

Mitchell_Porter25 Sep 2025 10:16 UTC

11 points

3 comments3 min readLW link

Petrov Day at Lighthaven

jimrandomh25 Sep 2025 8:29 UTC

20 points

0 comments1 min readLW link

Some Thoughts on Mech Interp

d4hines25 Sep 2025 6:10 UTC

13 points

1 comment8 min readLW link

IABIED is on the NYT bestseller list

Alice Blair25 Sep 2025 2:32 UTC

124 points

5 comments1 min readLW link

AI and the Hidden Price of Comfort

nickgpop25 Sep 2025 2:16 UTC

6 points

8 comments7 min readLW link

Nate Soares — If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All—with Jon Wolfsthal — at The Wharf

habryka25 Sep 2025 0:53 UTC

11 points

0 comments2 min readLW link

AGI Companies Won’t Profit From AGI

LTM24 Sep 2025 22:04 UTC

4 points

7 comments7 min readLW link

(routecause.substack.com)

Scheming Toy Environment: “Incompetent Client”

Ariel_24 Sep 2025 21:03 UTC

17 points

2 comments32 min readLW link

Synthesizing Standalone World-Models, Part 2: Shifting Structures

Thane Ruthenis24 Sep 2025 19:02 UTC

16 points

5 comments10 min readLW link

Alibaba won the AI wars, we just don’t see it yet

Misha Ramendik24 Sep 2025 18:45 UTC

−10 points

0 comments2 min readLW link

The Autofac Era

Gordon Seidoh Worley24 Sep 2025 18:20 UTC

29 points

18 comments7 min readLW link

(uncertainupdates.substack.com)

“Shut It Down” is simpler than “Controlled Takeoff”

Raemon24 Sep 2025 17:21 UTC

97 points

29 comments5 min readLW link

AISN #63: California’s SB-53 Passes the Legislature

Corin Katzke and Dan H

24 Sep 2025 17:02 UTC

6 points

0 comments4 min readLW link

(newsletter.safe.ai)

OpenAI Shows Us The Money

Zvi24 Sep 2025 15:30 UTC

40 points

8 comments9 min readLW link

(thezvi.wordpress.com)

Launching the $10,000 Existential Hope Meme Prize

elte24 Sep 2025 15:00 UTC

8 points

3 comments1 min readLW link

The Chinese Room re-visited: How LLM’s have real (but different) understanding of words

James Diacoumis24 Sep 2025 14:06 UTC

6 points

0 comments9 min readLW link

(jamesdiacoumis.substack.com)

An argument for discussing AI safety in person being underused

Kabir Kumar24 Sep 2025 11:36 UTC

17 points

1 comment2 min readLW link

How singleton contradicts longtermism

kapedalex24 Sep 2025 11:10 UTC

3 points

1 comment1 min readLW link

Berkeley Petrov Day

Darmani24 Sep 2025 7:59 UTC

6 points

0 comments1 min readLW link

EU and Monopoly on Violence

Martin Sustrik24 Sep 2025 7:51 UTC

118 points

3 comments5 min readLW link

(www.250bpm.com)

Misalignment and Roleplaying: Are Misaligned LLMs Acting Out Sci-Fi Stories?

Mark Keavney24 Sep 2025 2:09 UTC

30 points

4 comments13 min readLW link

A Possible Future: Decentralized AGI Proliferation

Dev.Errata23 Sep 2025 22:24 UTC

11 points

7 comments2 min readLW link

Munich, Bavaria “If Anyone Builds It” reading group

hilll23 Sep 2025 22:03 UTC

11 points

0 comments1 min readLW link

Prague “If Anyone Builds It” reading group

Marek Dědič23 Sep 2025 21:49 UTC

14 points

0 comments1 min readLW link

Draconian measures can increase the risk of irrevocable catastrophe

dsj23 Sep 2025 21:40 UTC

22 points

2 comments2 min readLW link

(thedavidsj.substack.com)

[Question] What the discontinuity is, if not FOOM?

TAG23 Sep 2025 19:30 UTC

18 points

14 comments3 min readLW link

Samuel Shadrach Interviewed

samuelshadrach23 Sep 2025 18:58 UTC

9 points

0 comments1 min readLW link

Statement of Support for “If Anyone Builds It, Everyone Dies”

Liron23 Sep 2025 17:51 UTC

67 points

34 comments1 min readLW link

Notes on fatalities from AI takeover

ryan_greenblatt23 Sep 2025 17:18 UTC

55 points

60 comments8 min readLW link

Zendo for large groups

philh23 Sep 2025 17:10 UTC

13 points

1 comment1 min readLW link

(reasonableapproximation.net)

Synthesizing Standalone World-Models, Part 1: Abstraction Hierarchies

Thane Ruthenis23 Sep 2025 17:01 UTC

23 points

10 comments23 min readLW link

A Compatibilist Definition of Santa Claus

Shiva's Right Foot23 Sep 2025 16:57 UTC

18 points

9 comments1 min readLW link

Ethics-Based Refusals Without Ethics-Based Refusal Training

1a3orn23 Sep 2025 16:35 UTC

91 points

2 comments19 min readLW link

Why Smarter Doesn’t Mean Kinder: Orthogonality and Instrumental Convergence

Alexander Müller23 Sep 2025 16:06 UTC

6 points

0 comments6 min readLW link

More Reactions to If Anyone Builds It, Everyone Dies

Zvi23 Sep 2025 16:00 UTC

33 points

20 comments20 min readLW link

(thezvi.wordpress.com)

Ontological Cluelessness

niplav and Claude+

23 Sep 2025 14:31 UTC

14 points

12 comments4 min readLW link

We are likely in an AI overhang, and this is bad.

Gabriel Alfour23 Sep 2025 14:15 UTC

55 points

16 comments1 min readLW link

(cognition.cafe)

Prompt optimization can enable AI control research

Mia Hopman and ZachParent

23 Sep 2025 12:46 UTC

35 points

3 comments9 min readLW link

Two Mathematical Perspectives on AI Hallucinations and Uncertainty

LorenzoPacchiardi23 Sep 2025 11:06 UTC

0 points

1 comment3 min readLW link

Accelerando as a “Slow, Reasonably Nice Takeoff” Story

Raemon23 Sep 2025 2:15 UTC

71 points

19 comments30 min readLW link

On failure, and keeping doors open; closing thoughts

jimmy23 Sep 2025 1:11 UTC

7 points

0 comments10 min readLW link