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

All Jan FebMarApr May Jun

All 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 192021 22 23 24 25 26 27 28 29 30 31

Improved visualizations of METR Time Horizons paper.

LDJMar 19, 2025, 11:36 PM

20 points

4 comments2 min readLW link

Is CCP authoritarianism good for building safe AI?

HrussMar 19, 2025, 11:13 PM

1 point

0 comments1 min readLW link

The case against “The case against AI alignment”

KvmanThinkingMar 19, 2025, 10:40 PM

2 points

0 comments1 min readLW link

[Question] Superintelligence Strategy: A Pragmatic Path to… Doom?

Mr BeastlyMar 19, 2025, 10:30 PM

6 points

0 comments3 min readLW link

SHIFT relies on token-level features to de-bias Bias in Bios probes

Tim HuaMar 19, 2025, 9:29 PM

39 points

2 comments6 min readLW link

Janet must die

ShmiMar 19, 2025, 8:35 PM

12 points

3 comments2 min readLW link

[Question] Why am I getting downvoted on Lesswrong?

OxidizeMar 19, 2025, 6:32 PM

7 points

14 comments1 min readLW link

Forecasting AI Futures Resource Hub

Alvin ÅnestrandMar 19, 2025, 5:26 PM

2 points

0 comments2 min readLW link

(forecastingaifutures.substack.com)

TBC episode w Dave Kasten from Control AI on AI Policy

EneaszMar 19, 2025, 5:09 PM

8 points

0 comments1 min readLW link

(www.thebayesianconspiracy.com)

Prioritizing threats for AI control

ryan_greenblattMar 19, 2025, 5:09 PM

58 points

2 comments10 min readLW link

The Illusion of Transparency as a Trust-Building Mechanism

Priyanka BharadwajMar 19, 2025, 5:09 PM

2 points

0 comments1 min readLW link

How Do We Govern AI Well?

kaimeMar 19, 2025, 5:08 PM

2 points

0 comments25 min readLW link

METR: Measuring AI Ability to Complete Long Tasks

Zach Stein-PerlmanMar 19, 2025, 4:00 PM

241 points

105 comments5 min readLW link

(metr.org)

Why I think AI will go poorly for humanity

Alek WestoverMar 19, 2025, 3:52 PM

13 points

0 comments30 min readLW link

The principle of genomic liberty

TsviBTMar 19, 2025, 2:27 PM

76 points

51 comments17 min readLW link

Going Nova

ZviMar 19, 2025, 1:30 PM

64 points

14 comments15 min readLW link

(thezvi.wordpress.com)

Equations Mean Things

abstractapplicMar 19, 2025, 8:16 AM

46 points

10 comments3 min readLW link

Elite Coordination via the Consensus of Power

Richard_NgoMar 19, 2025, 6:56 AM

92 points

15 comments12 min readLW link

(www.mindthefuture.info)

What I am working on right now and why: representation engineering edition

Lukasz G BartoszczeMar 18, 2025, 10:37 PM

3 points

0 comments3 min readLW link

Boots theory and Sybil Ramkin

philhMar 18, 2025, 10:10 PM

37 points

17 comments11 min readLW link

(reasonableapproximation.net)

Schmidt Sciences Technical AI Safety RFP on Inference-Time Compute – Deadline: April 30

Ryan GajarawalaMar 18, 2025, 6:05 PM

18 points

0 comments2 min readLW link

(www.schmidtsciences.org)

PRISM: Perspective Reasoning for Integrated Synthesis and Mediation (Interactive Demo)

Anthony DiamondMar 18, 2025, 6:03 PM

10 points

2 comments1 min readLW link

Subspace Rerouting: Using Mechanistic Interpretability to Craft Adversarial Attacks against Large Language Models

Le magicien quantiqueMar 18, 2025, 5:55 PM

6 points

1 comment10 min readLW link

Progress links and short notes, 2025-03-18

jasoncrawfordMar 18, 2025, 5:14 PM

8 points

0 comments3 min readLW link

(newsletter.rootsofprogress.org)

The Convergent Path to the Stars

Maxime RichéMar 18, 2025, 5:09 PM

6 points

0 comments20 min readLW link

Sapir-Whorf Ego Death

Jonathan MoregårdMar 18, 2025, 4:57 PM

8 points

7 comments2 min readLW link

(honestliving.substack.com)

Smelling Nice is Good, Actually

Gordon Seidoh WorleyMar 18, 2025, 4:54 PM

28 points

8 comments3 min readLW link

(uncertainupdates.substack.com)

A Taxonomy of Jobs Deeply Resistant to TAI Automation

Deric ChengMar 18, 2025, 4:25 PM

9 points

0 comments12 min readLW link

(www.convergenceanalysis.org)

Why Are The Human Sciences Hard? Two New Hypotheses

Aydin Mohseni, Daniel Herrmann and ben_levinstein

Mar 18, 2025, 3:45 PM

39 points

14 comments9 min readLW link

Go home GPT-4o, you’re drunk: emergent misalignment as lowered inhibitions

Stuart_Armstrong and rgorman

Mar 18, 2025, 2:48 PM

80 points

12 comments5 min readLW link

[Question] What is the theory of change behind writing papers about AI safety?

KajusMar 18, 2025, 12:51 PM

7 points

1 comment1 min readLW link

OpenAI #11: America Action Plan

ZviMar 18, 2025, 12:50 PM

83 points

3 comments6 min readLW link

(thezvi.wordpress.com)

I changed my mind about orca intelligence

Towards_KeeperhoodMar 18, 2025, 10:15 AM

46 points

24 comments5 min readLW link

[Question] Is Peano arithmetic trying to kill us? Do we care?

Q HomeMar 18, 2025, 8:22 AM

17 points

2 comments2 min readLW link

Do What the Mammals Do

CrimsonChinMar 18, 2025, 3:57 AM

2 points

6 comments4 min readLW link

What Actually Matters Until We Reach the Singularity

LexiusMar 18, 2025, 2:17 AM

−1 points

0 comments9 min readLW link

Meaning as a cognitive substitute for survival instincts: A thought experiment

Ovidijus ŠimkusMar 18, 2025, 1:53 AM

0 points

0 comments2 min readLW link

Against Yudkowsky’s evolution analogy for AI x-risk [unfinished]

Fiora SunshineMar 18, 2025, 1:41 AM

50 points

18 comments11 min readLW link

An “AI researcher” has written a paper on optimizing AI architecture and optimized a language model to several orders of magnitude more efficiency.

Y BMar 18, 2025, 1:15 AM

3 points

1 comment1 min readLW link

LessOnline 2025: Early Bird Tickets On Sale

Ben PaceMar 18, 2025, 12:22 AM

37 points

5 comments5 min readLW link

Feedback loops for exercise (VO2Max)

ElizabethMar 18, 2025, 12:10 AM

63 points

12 comments8 min readLW link

(acesounderglass.com)

FrontierMath Score of o3-mini Much Lower Than Claimed

YafahEdelmanMar 17, 2025, 10:41 PM

61 points

7 comments1 min readLW link

Proof-of-Concept Debugger for a Small LLM

Peter Lai and StefanHex

Mar 17, 2025, 10:27 PM

27 points

0 comments11 min readLW link

Effectively Communicating with DC Policymakers

PolicyTakesMar 17, 2025, 10:11 PM

14 points

0 comments2 min readLW link

Mind the Gap

Bridgett KayMar 17, 2025, 9:59 PM

8 points

0 comments5 min readLW link

(dxmrevealed.wordpress.com)

EIS XV: A New Proof of Concept for Useful Interpretability

scasperMar 17, 2025, 8:05 PM

30 points

2 comments3 min readLW link

Sentinel’s Global Risks Weekly Roundup #11/2025. Trump invokes Alien Enemies Act, Chinese invasion barges deployed in exercise.

NunoSempereMar 17, 2025, 7:34 PM

59 points

3 comments6 min readLW link

(blog.sentinel-team.org)

Claude Sonnet 3.7 (often) knows when it’s in alignment evaluations

Nicholas Goldowsky-Dill, Mikita Balesni, Jérémy Scheurer and Marius Hobbhahn

17 Mar 2025 19:11 UTC

182 points

9 comments6 min readLW link

Things Look Bleak for White-Collar Jobs Due to AI Acceleration

Declan Molony17 Mar 2025 17:03 UTC

15 points

0 comments10 min readLW link

Three Types of Intelligence Explosion

rosehadshar, Tom Davidson and wdmacaskill

17 Mar 2025 14:47 UTC

39 points

8 comments3 min readLW link

(www.forethought.org)