All 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 202120222023 2024 2025

All Jan Feb Mar Apr May Jun Jul Aug SepOctNov Dec

All 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 202122 23 24 25 26 27 28 29 30 31

22: Wait, We Did WHAT?

Zvi20 Oct 2022 21:50 UTC

55 points

16 comments16 min readLW link

(thezvi.wordpress.com)

When apparently positive evidence can be negative evidence

cata20 Oct 2022 21:47 UTC

19 points

5 comments1 min readLW link

(www.ncbi.nlm.nih.gov)

Plans Are Predictions, Not Optimization Targets

johnswentworth20 Oct 2022 21:17 UTC

108 points

20 comments4 min readLW link 1 review

Introduction to abstract entropy

Alex_Altair20 Oct 2022 21:03 UTC

234 points

78 comments18 min readLW link 1 review

Trajectories to 2036

ukc1001420 Oct 2022 20:23 UTC

3 points

1 comment14 min readLW link

[Question] Rough Sketch for Product to Enhance Citizen Participation in Politics

Fer32dwt34r3dfsz20 Oct 2022 20:04 UTC

13 points

5 comments1 min readLW link

The heritability of human values: A behavior genetic critique of Shard Theory

geoffreymiller20 Oct 2022 15:51 UTC

80 points

63 comments21 min readLW link

A Longtermist case against Veganism

Connor Tabarrok20 Oct 2022 14:30 UTC

−3 points

3 comments1 min readLW link

AI Research Program Prediction Markets

tailcalled20 Oct 2022 13:42 UTC

38 points

10 comments1 min readLW link

[Question] Is the meaning of words chosen/interpreted to maximize correlations with other relevant queries?

tailcalled20 Oct 2022 10:03 UTC

9 points

9 comments1 min readLW link

How to Write Readable Posts

David Hartsough20 Oct 2022 7:48 UTC

7 points

0 comments1 min readLW link

Notes on “Can you control the past”

So8res20 Oct 2022 3:41 UTC

64 points

41 comments21 min readLW link

Rhythmic Baby Toys

jefftk20 Oct 2022 1:50 UTC

15 points

1 comment1 min readLW link

(www.jefftk.com)

[Question] What Does AI Alignment Success Look Like?

Shmi20 Oct 2022 0:32 UTC

23 points

7 comments1 min readLW link

Scaling Laws for Reward Model Overoptimization

leogao, John Schulman and Jacob_Hilton

20 Oct 2022 0:20 UTC

103 points

13 comments1 min readLW link

(arxiv.org)

What is Consciousness?

belkarx19 Oct 2022 21:14 UTC

3 points

2 comments2 min readLW link

What to do if a nuclear weapon is used in Ukraine?

Valentin202619 Oct 2022 18:43 UTC

13 points

9 comments3 min readLW link

[Question] If I asked for an explanation of a perfect Utopia, could you give one?

Akkira19 Oct 2022 17:56 UTC

−4 points

2 comments1 min readLW link

[Question] Should we push for requiring AI training data to be licensed?

ChristianKl19 Oct 2022 17:49 UTC

37 points

32 comments1 min readLW link

Hacker-AI and Digital Ghosts – Pre-AGI

Erland Wittkotter19 Oct 2022 15:33 UTC

9 points

7 comments8 min readLW link

The reward function is already how well you manipulate humans

Kerry19 Oct 2022 1:52 UTC

20 points

9 comments2 min readLW link

Response to Katja Grace’s AI x-risk counterarguments

Erik Jenner and Johannes Treutlein

19 Oct 2022 1:17 UTC

77 points

18 comments15 min readLW link

(OLD) An Extremely Opinionated Annotated List of My Favourite Mechanistic Interpretability Papers

Neel Nanda18 Oct 2022 21:08 UTC

72 points

5 comments12 min readLW link

(www.neelnanda.io)

Distilled Representations Research Agenda

Hoagy and mishajw

18 Oct 2022 20:59 UTC

15 points

2 comments8 min readLW link

Drafting a Covid Survey

jefftk18 Oct 2022 19:30 UTC

15 points

2 comments2 min readLW link

(www.jefftk.com)

How To Make Prediction Markets Useful For Alignment Work

johnswentworth18 Oct 2022 19:01 UTC

97 points

18 comments2 min readLW link

A conversation about Katja’s counterarguments to AI risk

Matthew Barnett and Ege Erdil

18 Oct 2022 18:40 UTC

43 points

9 comments33 min readLW link

ACX Zurich October Meetup

MB18 Oct 2022 18:24 UTC

1 point

1 comment1 min readLW link

Untapped Potential at 13-18

belkarx18 Oct 2022 18:09 UTC

82 points

53 comments1 min readLW link

[Question] How easy is it to supervise processes vs outcomes?

Noosphere8918 Oct 2022 17:48 UTC

3 points

0 comments1 min readLW link

Is GitHub Copilot in legal trouble?

tcelferact18 Oct 2022 16:19 UTC

35 points

2 comments1 min readLW link

Metaculus is building a team dedicated to AI forecasting

ChristianWilliams18 Oct 2022 16:08 UTC

3 points

0 comments1 min readLW link

How to Take Over the Universe (in Three Easy Steps)

Writer18 Oct 2022 15:04 UTC

47 points

17 comments12 min readLW link

(youtu.be)

Science of Deep Learning—a technical agenda

Marius Hobbhahn18 Oct 2022 14:54 UTC

36 points

7 comments4 min readLW link

My search for a reliable breakfast

tomdekan18 Oct 2022 9:42 UTC

6 points

17 comments3 min readLW link

(www.tomdekan.com)

Infinite Possibility Space and the Shutdown Problem

magfrump18 Oct 2022 5:37 UTC

9 points

0 comments2 min readLW link

(www.magfrump.net)

Audition to perform in Bay Secular Solstice

mingyuan18 Oct 2022 3:10 UTC

25 points

3 comments1 min readLW link

Decision theory does not imply that we get to have nice things

So8res18 Oct 2022 3:04 UTC

171 points

73 comments26 min readLW link 2 reviews

South Bay ACX/LW Meetup

IS18 Oct 2022 2:50 UTC

2 points

1 comment1 min readLW link

Vertical Flutes

jefftk18 Oct 2022 1:40 UTC

10 points

4 comments1 min readLW link

(www.jefftk.com)

Why Weren’t Hot Air Balloons Invented Sooner?

Lost Futures18 Oct 2022 0:41 UTC

115 points

52 comments6 min readLW link

(lostfutures.substack.com)

Is GPT-N bounded by human capabilities? No.

Cleo Nardo17 Oct 2022 23:26 UTC

49 points

8 comments2 min readLW link

EA & LW Forums Weekly Summary (10 − 16 Oct 22′)

Zoe Williams17 Oct 2022 22:51 UTC

12 points

4 comments1 min readLW link

A pragmatic metric for Artificial General Intelligence

lorepieri17 Oct 2022 22:07 UTC

6 points

0 comments1 min readLW link

(lorenzopieri.com)

They gave LLMs access to physics simulators

ryan_b17 Oct 2022 21:21 UTC

50 points

18 comments1 min readLW link

(arxiv.org)

Combatting perfectionism

tomdekan17 Oct 2022 20:58 UTC

6 points

0 comments2 min readLW link

(tomdekan.com)

Open Problem in Voting Theory

Scott Garrabrant17 Oct 2022 20:42 UTC

75 points

16 comments6 min readLW link

Maximal Lottery-Lotteries

Scott Garrabrant17 Oct 2022 20:39 UTC

72 points

15 comments4 min readLW link

[Question] Creating superintelligence without AGI

Antb17 Oct 2022 19:01 UTC

7 points

3 comments1 min readLW link

AI Safety Ideas: An Open AI Safety Research Platform

Esben Kran17 Oct 2022 17:01 UTC

24 points

0 comments1 min readLW link