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 JunJulAug 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 242526 27 28 29 30 31

The Long (Self-)Correction

Wei Dai24 Jul 2026 21:01 UTC

268 points

62 comments2 min readLW link

Why PauseAI UK accepts anonymous donations

Joseph Miller and PauseAI UK

24 Jul 2026 20:37 UTC

13 points

0 comments5 min readLW link

Seeking Mentees for the Sentient Futures Project Incubator

BrodyM24 Jul 2026 19:52 UTC

7 points

0 comments1 min readLW link

Intent Is All You Need.

Not Sure24 Jul 2026 19:52 UTC

−7 points

2 comments2 min readLW link

Congress Moves at Tech Pace: The FRONTIER Act

dan.parshall24 Jul 2026 19:34 UTC

26 points

0 comments8 min readLW link

Stable Systems Have Stable Outputs

Deixis24 Jul 2026 19:21 UTC

5 points

0 comments4 min readLW link

The AI Industrial Explosion — Part 5: Given AGI, automating physical production is probably not that hard

djbinder24 Jul 2026 19:15 UTC

29 points

2 comments25 min readLW link

(defensesindepth.bio)

Where does hint-following and concealment arise? A case study on OLMo-3 checkpoints

arav-dhoot and yix

24 Jul 2026 19:14 UTC

24 points

0 comments4 min readLW link

Introducing Lightcone Commons

Zvi24 Jul 2026 17:21 UTC

35 points

0 comments6 min readLW link

(thezvi.wordpress.com)

Should we be worried about how good AI is getting at coding autonomous drones?

Lukas Petersson24 Jul 2026 16:44 UTC

6 points

9 comments1 min readLW link

Coefficient Giving just gave GiveWell $1 billion. Where should other donors give now?

jackultraphil24 Jul 2026 15:40 UTC

4 points

0 comments2 min readLW link

(fundinganthropalypse.com)

LLMs are (still) mostly powered by imitative learning, not RL

Steven Byrnes24 Jul 2026 14:26 UTC

239 points

39 comments9 min readLW link

Democracy isn’t ready for the AI revolution

Sophia Gore24 Jul 2026 14:17 UTC

25 points

3 comments5 min readLW link

Does distilling Claude carry the persona with it?

Benji Berczi and Kyuhee Kim

24 Jul 2026 12:31 UTC

43 points

4 comments10 min readLW link

Georgia Tech AI Safety Initiative Retrospective 2025-2026

Ishan Khire, yix, Andersehen, Alec Harris, afterless, Eyas Ayesh, RocioPV and hersheys

24 Jul 2026 11:55 UTC

61 points

3 comments7 min readLW link

[Linkpost] Thoughts on the Recent OpenAI Hack

Linch24 Jul 2026 1:51 UTC

19 points

2 comments4 min readLW link

Should OpenAI’s rogue agent be punished?

groblegark24 Jul 2026 1:20 UTC

1 point

3 comments1 min readLW link

Evaluating Red Team and Blue Team Capability for AI Control Research

Ram Potham24 Jul 2026 1:11 UTC

11 points

0 comments8 min readLW link

(dearfutureais.substack.com)

Fixing rewards for NLA to reduce confabulation

SEONG PYO HONG24 Jul 2026 0:55 UTC

8 points

0 comments4 min readLW link

Anthropic’s J-Lens: A Research Engineer’s Analysis

willkn24 Jul 2026 0:54 UTC

8 points

0 comments11 min readLW link

Pulling the Fire Alarm

nem23 Jul 2026 23:14 UTC

86 points

5 comments1 min readLW link

Contra George Hotz on “AI 2040 and the Cult of Intelligence”

Matthew Tromp23 Jul 2026 22:43 UTC

8 points

0 comments8 min readLW link

(substack.com)

The Model Organism Lottery: Model Organism Interpretability Strongly Depends on Training Methodology

GabrielKS, therootof3, Raffaello Fornasiere, Nikita Menon and StefanHex

23 Jul 2026 22:37 UTC

45 points

0 comments6 min readLW link

(arxiv.org)

vibes-based thinking as a cultural response to unknowns

madelineberzak23 Jul 2026 22:35 UTC

7 points

0 comments8 min readLW link

Why an LLM cannot accumulate concepts

Zenya23 Jul 2026 21:40 UTC

−10 points

0 comments2 min readLW link

Red-teaming LLM unlearning: LUNAR’s “forgotten” knowledge is still recoverable

Diksha Gupta23 Jul 2026 21:39 UTC

10 points

0 comments5 min readLW link

Inception in DiffusionGemma—Jailbreaking a Diffusion Language Model by Pinning Tokens Anywhere on the Canvas

Theresa G., Simon S. and Siva Kumar Lakkoju

23 Jul 2026 21:39 UTC

19 points

0 comments11 min readLW link

ACX Atlanta August Meetup

Steve French23 Jul 2026 21:31 UTC

2 points

0 comments1 min readLW link

Not Pinning Your OpenRouter Provider Might Invalidate Your Research

Matthew Khoriaty23 Jul 2026 20:17 UTC

112 points

16 comments8 min readLW link

Pseudpocalypse

dynomight23 Jul 2026 20:09 UTC

60 points

18 comments23 min readLW link

Estimating LLM Training FLOPs on the Nvidia Jetson Orin Nano

William Fowler23 Jul 2026 19:52 UTC

24 points

0 comments10 min readLW link

Lightcone Commons

habryka23 Jul 2026 18:35 UTC

377 points

30 comments19 min readLW link

Challenge: Hand coding weights for efficient sequence memorisation

Linda Linsefors and Lucius Bushnaq

23 Jul 2026 18:05 UTC

65 points

9 comments18 min readLW link

AI Researchers Don’t Understand the State

Alex Amadori23 Jul 2026 17:58 UTC

3 points

0 comments1 min readLW link

(x.com)

The OpenAI/Huggingface incident | Redwood Research podcast episode 2

ryan_greenblatt and Buck

23 Jul 2026 17:56 UTC

85 points

0 comments2 min readLW link

V&V takes on OpenAI’s long-horizon incidents

Yoav Hollander23 Jul 2026 16:51 UTC

22 points

0 comments4 min readLW link

(blog.foretellix.com)

Duane Arnold

Tomás B.23 Jul 2026 16:17 UTC

138 points

4 comments19 min readLW link

Introducing Impact List: a ranking of people by the expected value of their donations

Elliot_Olds23 Jul 2026 15:27 UTC

12 points

0 comments5 min readLW link

Wanting Crooked Lines

Davey Morse23 Jul 2026 14:33 UTC

13 points

0 comments4 min readLW link

(3/3) The Dangers of ASI

Eigenbraid23 Jul 2026 13:32 UTC

10 points

0 comments3 min readLW link

AI #178: A Fire Alarm For General Intelligence

Zvi23 Jul 2026 13:21 UTC

41 points

1 comment43 min readLW link

(thezvi.wordpress.com)

Mathematicians are Feeling the Doom

alkjash23 Jul 2026 13:16 UTC

192 points

54 comments1 min readLW link

Sleeping Beauty as a Mind Killer

avturchin23 Jul 2026 11:04 UTC

9 points

25 comments7 min readLW link

Get You a Model: If You’re Not Paying, You’re Missing 90% Of Improvements

Czynski23 Jul 2026 4:16 UTC

7 points

5 comments1 min readLW link

(dangeroussincerity.substack.com)

Are we existentially threatened by the type of AI misalignment seen in the OpenAI Hugging Face attack?

Alex Mallen and Girish Gupta

23 Jul 2026 3:40 UTC

243 points

12 comments5 min readLW link

Necessity Protects Chain of Thought Monitoring by Prevention, Not Disclosure

Ankit Sanjyal23 Jul 2026 2:51 UTC

7 points

0 comments10 min readLW link

After 3000 hours of meditation I’m convinced multi-agent models of mind are just true

Arram Sabeti23 Jul 2026 2:50 UTC

2 points

4 comments1 min readLW link

(arram.substack.com)

LLMOSES

Davey Morse23 Jul 2026 2:35 UTC

−5 points

0 comments3 min readLW link

Reading Dudley and Jacqueline’s Barn Dance Book

jefftk23 Jul 2026 2:21 UTC

12 points

3 comments3 min readLW link

(www.jefftk.com)

A Multi-Agent Extension for Petri

carissacullen22 Jul 2026 21:51 UTC

11 points

0 comments4 min readLW link