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 212223 24 25 26 27 28 29 30

You Can’t Really Bet on Doom

Jack_S21 Sep 2025 23:27 UTC

8 points

1 comment7 min readLW link

(torchestogether.substack.com)

The Only Red Line

Jason Reid21 Sep 2025 22:40 UTC

13 points

1 comment1 min readLW link

Do LLMs Change Their Minds About Their Users… and Know It?

Ishaan Sinha21 Sep 2025 22:38 UTC

10 points

2 comments14 min readLW link

Metacrisis as a Framework for AI Governance

Jonah Wilberg21 Sep 2025 21:30 UTC

20 points

0 comments8 min readLW link

Is there not legitimate disagreement about this premise of IABI,ED?

enfascination21 Sep 2025 20:47 UTC

5 points

7 comments1 min readLW link

Evals in the Age of Jarvis

Dinkar Juyal21 Sep 2025 19:27 UTC

3 points

2 comments3 min readLW link

[Question] Could China Unilaterally Cause an AI Pause?

Maloew21 Sep 2025 18:37 UTC

22 points

2 comments1 min readLW link

What do people mean when they say that something will become more like a utility maximizer?

Nina Panickssery21 Sep 2025 16:03 UTC

40 points

7 comments2 min readLW link

And Yet, Defend your Thoughts from AI Writing

Michael Samoilov21 Sep 2025 15:52 UTC

60 points

17 comments6 min readLW link

(open.substack.com)

A parable of realism and relativism

kwang21 Sep 2025 14:47 UTC

−7 points

2 comments2 min readLW link

(kevw.substack.com)

ACX/LW October Paris Meetup

Lucie Philippon21 Sep 2025 11:37 UTC

5 points

0 comments1 min readLW link

Day #8 Hunger Strike, Protest Against Superintelligent AI

samuelshadrach21 Sep 2025 5:58 UTC

13 points

4 comments2 min readLW link

FTX, Golden Geese, and The Widow’s Mite

Elizabeth20 Sep 2025 18:30 UTC

21 points

1 comment7 min readLW link

(acesounderglass.com)

The Case for a Pro-AI-Safety Political Party in the US

Oliver Kuperman20 Sep 2025 16:35 UTC

11 points

2 comments21 min readLW link

Contra Collier on IABIED

Max Harms20 Sep 2025 15:55 UTC

227 points

51 comments20 min readLW link

Astralcodexten IRB history error

Paul Crowley20 Sep 2025 15:28 UTC

36 points

0 comments2 min readLW link

“Pastor Selfie,” How would you teach what you learn about rationality from scratch to the general public?

P. João20 Sep 2025 12:43 UTC

1 point

1 comment1 min readLW link

The Problem with Defining an “AGI Ban” by Outcome (a lawyer’s take).

Katalina Hernandez20 Sep 2025 11:01 UTC

239 points

63 comments5 min readLW link

The title is reasonable

Raemon20 Sep 2025 8:59 UTC

194 points

128 comments18 min readLW link

An Economic Model of Modern Dating

gladman20 Sep 2025 2:17 UTC

4 points

0 comments4 min readLW link

Announcing “The Real AI”: a blog

David Scott Krueger (formerly: capybaralet)20 Sep 2025 1:27 UTC

32 points

1 comment2 min readLW link

(therealartificialintelligence.substack.com)

Extending Inspect Framework: Integrating Weights & Biases

Qi Guo, Matan Shtepel, Daniel Polatajko and Justin Olive

20 Sep 2025 1:10 UTC

2 points

0 comments3 min readLW link

[Question] Looking for a ray of hope in IABIED

Rich Mansfield20 Sep 2025 0:53 UTC

11 points

3 comments1 min readLW link

Memory Decoding Journal Club: Distinct synaptic plasticity rules operate across dendritic compartments in vivo during learning

Devin Ward20 Sep 2025 0:50 UTC

1 point

0 comments1 min readLW link

Beliefs and JavaScript types

Adam Zerner20 Sep 2025 0:48 UTC

10 points

6 comments6 min readLW link

AI Lobbying is Not Normal

Algon20 Sep 2025 0:23 UTC

131 points

11 comments3 min readLW link

(x.com)

Beware LLMs’ pathological guardrailing

lc19 Sep 2025 20:55 UTC

20 points

1 comment1 min readLW link

Safety researchers should take a public stance

Mateusz Bagiński and Ishual

19 Sep 2025 18:55 UTC

230 points

65 comments8 min readLW link

Day 16 Hunger Strike—Guido Reichstader Interviewed

samuelshadrach19 Sep 2025 17:30 UTC

9 points

0 comments1 min readLW link

Prospects for studying actual schemers

ryan_greenblatt and Julian Stastny

19 Sep 2025 14:11 UTC

40 points

0 comments58 min readLW link

Book Review: If Anyone Builds It, Everyone Dies

Zvi19 Sep 2025 11:30 UTC

61 points

3 comments31 min readLW link

(thezvi.wordpress.com)

How people politically confront the Modern Eldritch

PranavG and Gabriel Alfour

19 Sep 2025 10:18 UTC

5 points

0 comments14 min readLW link

(cognition.cafe)

My Minor AI Safety Research Projects (Q3 2025)

Adam Newgas19 Sep 2025 9:53 UTC

6 points

1 comment2 min readLW link

 Book Review: If Anyone Builds It, Everyone Dies

Nina Panickssery19 Sep 2025 4:50 UTC

41 points

1 comment11 min readLW link

(blog.ninapanickssery.com)

Memory Decoding Journal Club: Distinct synaptic plasticity rules operate across dendritic compartments in vivo during learning

Devin Ward19 Sep 2025 4:17 UTC

3 points

0 comments1 min readLW link

AI psychosis isn’t really psychosis

GGWG19 Sep 2025 3:18 UTC

5 points

1 comment1 min readLW link

JDP Reviews IABIED

jdp19 Sep 2025 1:23 UTC

74 points

21 comments8 min readLW link

(minihf.com)

Teaching My Toddler To Read

maia19 Sep 2025 0:17 UTC

155 points

17 comments10 min readLW link

IABIED Review—An Unfortunate Miss

Darren McKee18 Sep 2025 22:39 UTC

65 points

22 comments9 min readLW link

You can’t eval GPT5 anymore

Lukas Petersson18 Sep 2025 22:12 UTC

152 points

11 comments1 min readLW link

Oxford – ACX Meetups Everywhere Fall 2025

fenmund and Sam F. Brown

18 Sep 2025 20:22 UTC

1 point

0 comments1 min readLW link

If anyone builds it, everyone will plausibly be fine

joshc18 Sep 2025 20:03 UTC

29 points

24 comments7 min readLW link

It Never Worked Before: Nine Intellectual Jokes

Linch18 Sep 2025 19:48 UTC

13 points

2 comments2 min readLW link

(linch.substack.com)

An Attempt to Explain my AI Risk Explainer Attempt

thenoviceoof18 Sep 2025 19:35 UTC

11 points

2 comments10 min readLW link

(thenoviceoof.com)

More Was Possible: A Review of IABIED

Vaniver18 Sep 2025 19:33 UTC

54 points

5 comments1 min readLW link

(asteriskmag.com)

Can an AI become human?

Robert Shuler18 Sep 2025 19:18 UTC

3 points

0 comments19 min readLW link

The Strange Case of Emergent Misalignment

Alexander Müller and ilijalichkovski

18 Sep 2025 14:45 UTC

2 points

0 comments5 min readLW link

AI #134: If Anyone Reads It

Zvi18 Sep 2025 13:10 UTC

34 points

8 comments61 min readLW link

(thezvi.wordpress.com)

These are my reasons to worry less about loss of control over LLM-based agents

otto.barten18 Sep 2025 11:45 UTC

7 points

4 comments4 min readLW link

The End-of-the-World Party

Jakub Growiec18 Sep 2025 7:49 UTC

1 point

0 comments53 min readLW link