8 Jun 2023 23:40 UTC

72 points

3 comments22 min readLW link

Updates and Reflections on Optimal Exercise after Nearly a Decade

romeostevensit8 Jun 2023 23:02 UTC

187 points

55 comments2 min readLW link

Takeaways from the Mechanistic Interpretability Challenges

scasper8 Jun 2023 18:56 UTC

93 points

5 comments6 min readLW link

Leave an Emotional Line of Retreat

Johannes C. Mayer8 Jun 2023 18:36 UTC

23 points

1 comment1 min readLW link

Current AI harms are also sci-fi

Christopher King8 Jun 2023 17:49 UTC

26 points

3 comments1 min readLW link

Two Ways To Reduce Unhappiness That Comes From Distorted Views of Reality

Anne Hsu8 Jun 2023 17:43 UTC

3 points

0 comments7 min readLW link

Collaboration in Science: Happier People ↔ Better Research

nadinespy8 Jun 2023 17:42 UTC

3 points

0 comments32 min readLW link

Biomimetic alignment: Alignment between animal genes and animal brains as a model for alignment between humans and AI systems

geoffreymiller8 Jun 2023 16:05 UTC

10 points

0 comments16 min readLW link

A potentially high impact differential technological development area

Noosphere898 Jun 2023 14:33 UTC

5 points

2 comments2 min readLW link

[Question] Question for Prediction Market people: where is the money supposed to come from?

Robert_AIZI8 Jun 2023 13:58 UTC

25 points

26 comments1 min readLW link

AI #15: The Principle of Charity

Zvi8 Jun 2023 12:10 UTC

73 points

16 comments44 min readLW link

(thezvi.wordpress.com)

if you’re reading this it’s too late (a new theory on what is causing the Great Stagnation)

rogersbacon8 Jun 2023 11:49 UTC

−10 points

2 comments13 min readLW link

(www.secretorum.life)

[Linkpost] Scaling laws for language encoding models in fMRI

Bogdan Ionut Cirstea8 Jun 2023 10:52 UTC

30 points

0 comments1 min readLW link

Transformative AI is a process

meijer19738 Jun 2023 8:57 UTC

2 points

0 comments5 min readLW link

Crisis of Faith case study: beyond reductionism?

MalcolmOcean8 Jun 2023 6:11 UTC

6 points

9 comments19 min readLW link

I wrote this because of watermelon

Arti8 Jun 2023 3:55 UTC

2 points

2 comments1 min readLW link

Learning Transformer Programs [Linkpost]

aogara8 Jun 2023 0:16 UTC

7 points

0 comments1 min readLW link

(arxiv.org)

What will GPT-2030 look like?

jsteinhardt7 Jun 2023 23:40 UTC

182 points

42 comments23 min readLW link

(bounded-regret.ghost.io)

Progress links and tweets, 2023-06-07

jasoncrawford7 Jun 2023 23:26 UTC

11 points

0 comments1 min readLW link

(rootsofprogress.org)

LEAst-squares Concept Erasure (LEACE)

tricky_labyrinth7 Jun 2023 21:51 UTC

68 points

10 comments1 min readLW link

(twitter.com)

Proposal: Tune LLMs to Use Calibrated Language

OneManyNone7 Jun 2023 21:05 UTC

9 points

0 comments5 min readLW link

A moral backlash against AI will probably slow down AGI development

geoffreymiller7 Jun 2023 20:39 UTC

49 points

10 comments14 min readLW link

RAMP—RoboNet Artificial Media Protocol

antoniomax7 Jun 2023 19:01 UTC

−1 points

0 comments19 min readLW link

(antoniomax.substack.com)

An Exercise to Build Intuitions on AGI Risk

Lauro Langosco7 Jun 2023 18:35 UTC

52 points

3 comments8 min readLW link

Elon talked with senior Chinese leadership about AI X-risk

ChristianKl7 Jun 2023 15:02 UTC

47 points

2 comments1 min readLW link

(www.youtube.com)

Article Summary: Current and Near-Term AI as a Potential Existential Risk Factor

André Ferretti7 Jun 2023 13:51 UTC

28 points

3 comments1 min readLW link

(dl.acm.org)

gamers beware: modded Minecraft has new malware

the gears to ascension7 Jun 2023 13:49 UTC

14 points

5 comments1 min readLW link

(github.com)

Launching Lightspeed Grants (Apply by July 6th)

habryka7 Jun 2023 2:53 UTC

211 points

41 comments5 min readLW link

Cultivate an obsession with the object level

Richard_Ngo7 Jun 2023 1:39 UTC

70 points

4 comments3 min readLW link

How to Slow AI Development

PeterMcCluskey7 Jun 2023 0:29 UTC

20 points

0 comments5 min readLW link

(bayesianinvestor.com)

[Question] Killing Recurrent Memory Over Self Attention?

Del Nobolo6 Jun 2023 23:02 UTC

3 points

0 comments1 min readLW link

[Job Ad] SERI MATS is (still) hiring for our summer program

Ryan Kidd and zanekay

6 Jun 2023 21:07 UTC

12 points

0 comments7 min readLW link

Why I am not a longtermist (May 2022)

boazbarak6 Jun 2023 20:36 UTC

39 points

18 comments9 min readLW link

(windowsontheory.org)

Society Library seeking contributions for canonical AI Safety debate map

Jarred Filmer6 Jun 2023 18:15 UTC

36 points

0 comments1 min readLW link

(www.societylibrary.org)

A Playbook for AI Risk Reduction (focused on misaligned AI)

HoldenKarnofsky6 Jun 2023 18:05 UTC

90 points

41 comments14 min readLW link

A “bottom-up” approach to AI as a more transparent alternative to “top-down” LLMs

Paul Jorion6 Jun 2023 18:00 UTC

1 point

0 comments1 min readLW link

Why Yudkowsky Is Wrong And What He Does Can Be More Dangerous

idontagreewiththat6 Jun 2023 17:59 UTC

−40 points

3 comments3 min readLW link

The Base Rate Times, news through prediction markets

vandemonian6 Jun 2023 17:42 UTC

267 points

39 comments4 min readLW link

Monthly Roundup #7: June 2023

Zvi6 Jun 2023 17:40 UTC

23 points

13 comments43 min readLW link

(thezvi.wordpress.com)

Transformative AGI by 2043 is <1% likely

Ted Sanders6 Jun 2023 17:36 UTC

34 points

115 comments5 min readLW link

(arxiv.org)

AISN #9: Statement on Extinction Risks, Competitive Pressures, and When Will AI Reach Human-Level?

Dan H and aogara

6 Jun 2023 16:10 UTC

12 points

0 comments7 min readLW link

(newsletter.safe.ai)

An Eternal Company

moyamo6 Jun 2023 15:56 UTC

7 points

8 comments4 min readLW link

AISC end of program presentations

Linda Linsefors and Remmelt

6 Jun 2023 15:45 UTC

18 points

0 comments1 min readLW link

Why the Solutions to AI Alignment are Likely Outside the Overton Window

williamsae6 Jun 2023 14:21 UTC

−6 points

0 comments3 min readLW link

Stampy’s AI Safety Info—New Distillations #3 [May 2023]

markov6 Jun 2023 14:18 UTC

16 points

0 comments2 min readLW link

(aisafety.info)

Agentic Mess (A Failure Story)

Karl von Wendt, Sofia Bharadia, PeterDrotos, Artem Korotkov, mespa and mruwnik

6 Jun 2023 13:09 UTC

44 points

5 comments13 min readLW link

Berlin AI Alignment Open Meetup June 2023

GuyP6 Jun 2023 10:04 UTC

5 points

0 comments1 min readLW link

The Sharp Right Turn: sudden deceptive alignment as a convergent goal

avturchin6 Jun 2023 9:59 UTC

38 points

5 comments1 min readLW link

Open Thread: June 2023 (Inline Reacts!)

Raemon6 Jun 2023 7:40 UTC

19 points

57 comments1 min readLW link

[Linkpost] Given Extinction Worries, Why Don’t AI Researchers Quit? Well, Several Reasons

Daniel_Eth6 Jun 2023 7:31 UTC

10 points

0 comments1 min readLW link