All 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 202220232024 2025

All Jan Feb Mar AprMayJun Jul Aug Sep Oct Nov Dec

All 1 2 3 4 5 678 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Time and Energy Costs to Erase a Bit

DaemonicSigilMay 6, 2023, 11:29 PM

24 points

11 votes

Overall karma indicates overall quality.

32 comments7 min readLW link

How much do you believe your results?

Eric NeymanMay 6, 2023, 8:31 PM

514 points

235 votes

Overall karma indicates overall quality.

18 comments15 min readLW link 4 reviews

(ericneyman.wordpress.com)

Long Covid Risks: 2023 Update

ElizabethMay 6, 2023, 6:20 PM

69 points

31 votes

Overall karma indicates overall quality.

11 comments4 min readLW link

(acesounderglass.com)

Is “red” for GPT-4 the same as “red” for you?

Yusuke HayashiMay 6, 2023, 5:55 PM

9 points

8 votes

Overall karma indicates overall quality.

6 comments2 min readLW link

The Broader Fossil Fuel Community

Jeffrey HeningerMay 6, 2023, 2:49 PM

16 points

12 votes

Overall karma indicates overall quality.

1 comment3 min readLW link

Estimating Norovirus Prevalence

jefftkMay 6, 2023, 11:40 AM

16 points

4 votes

Overall karma indicates overall quality.

0 comments2 min readLW link

(www.jefftk.com)

Alignment as Function Fitting

A.H.May 6, 2023, 11:38 AM

7 points

3 votes

Overall karma indicates overall quality.

0 comments12 min readLW link

My preferred framings for reward misspecification and goal misgeneralisation

Yi-YangMay 6, 2023, 4:48 AM

27 points

10 votes

Overall karma indicates overall quality.

1 comment8 min readLW link

You don’t need to be a genius to be in AI safety research

Claire ShortMay 6, 2023, 2:32 AM

15 points

28 votes

Overall karma indicates overall quality.

1 comment6 min readLW link

Naturalist Collection

LoganStrohlMay 6, 2023, 12:37 AM

71 points

21 votes

Overall karma indicates overall quality.

7 comments15 min readLW link

Do you work at an AI lab? Please quit

Nik SamoylovMay 5, 2023, 11:41 PM

−29 points

13 votes

Overall karma indicates overall quality.

9 comments1 min readLW link

Explaining “Hell is Game Theory Folk Theorems”

electroswingMay 5, 2023, 11:33 PM

57 points

31 votes

Overall karma indicates overall quality.

21 comments5 min readLW link

Sleeping Beauty – the Death Hypothesis

Guillaume CharrierMay 5, 2023, 11:32 PM

7 points

10 votes

Overall karma indicates overall quality.

8 comments5 min readLW link

Orthogonal’s Formal-Goal Alignment theory of change

Tamsin LeakeMay 5, 2023, 10:36 PM

69 points

33 votes

Overall karma indicates overall quality.

13 comments4 min readLW link

(carado.moe)

A smart enough LLM might be deadly simply if you run it for long enough

Mikhail SaminMay 5, 2023, 8:49 PM

19 points

17 votes

Overall karma indicates overall quality.

16 comments8 min readLW link

What Jason has been reading, May 2023: “Protopia,” complex systems, Daedalus vs. Icarus, and more

jasoncrawfordMay 5, 2023, 7:54 PM

25 points

8 votes

Overall karma indicates overall quality.

2 comments11 min readLW link

(rootsofprogress.org)

CHAT Diplomacy: LLMs and National Security

SebastianG May 5, 2023, 7:45 PM

25 points

10 votes

Overall karma indicates overall quality.

6 comments7 min readLW link

Linkpost for Accursed Farms Discussion / debate with AI expert Eliezer Yudkowsky

gilchMay 5, 2023, 6:20 PM

14 points

9 votes

Overall karma indicates overall quality.

2 comments1 min readLW link

(www.youtube.com)

Regulate or Compete? The China Factor in U.S. AI Policy (NAIR #2)

charles_mMay 5, 2023, 5:43 PM

2 points

2 votes

Overall karma indicates overall quality.

1 comment7 min readLW link

(navigatingairisks.substack.com)

Kingfisher Live CD Process

jefftkMay 5, 2023, 5:00 PM

13 points

3 votes

Overall karma indicates overall quality.

0 comments3 min readLW link

(www.jefftk.com)

What can we learn from Bayes about reasoning?

jasoncrawfordMay 5, 2023, 3:52 PM

22 points

7 votes

Overall karma indicates overall quality.

11 comments1 min readLW link

[Question] Why not use active SETI to prevent AI Doom?

RomanSMay 5, 2023, 2:41 PM

13 points

20 votes

Overall karma indicates overall quality.

13 comments1 min readLW link

Investigating Emergent Goal-Like Behavior in Large Language Models using Experimental Economics

phelps-sgMay 5, 2023, 11:15 AM

6 points

4 votes

Overall karma indicates overall quality.

1 comment4 min readLW link

Monthly Shorts 4/23

CelerMay 5, 2023, 7:20 AM

8 points

4 votes

Overall karma indicates overall quality.

1 comment3 min readLW link

(keller.substack.com)

[Question] What is it like to be a compatibilist?

tslarmMay 5, 2023, 2:56 AM

8 points

5 votes

Overall karma indicates overall quality.

72 comments1 min readLW link

Transcript of a presentation on catastrophic risks from AI

RobertMMay 5, 2023, 1:38 AM

6 points

1 vote

Overall karma indicates overall quality.

0 comments8 min readLW link

How to get good at programming

Ulisse MiniMay 5, 2023, 1:14 AM

40 points

25 votes

Overall karma indicates overall quality.

3 comments2 min readLW link

A brief collection of Hinton’s recent comments on AGI risk

Kaj_SotalaMay 4, 2023, 11:31 PM

148 points

58 votes

Overall karma indicates overall quality.

9 comments11 min readLW link

Robin Hanson and I talk about AI risk

KatjaGraceMay 4, 2023, 10:20 PM

39 points

11 votes

Overall karma indicates overall quality.

8 comments1 min readLW link

(worldspiritsockpuppet.com)

Who regulates the regulators? We need to go beyond the review-and-approval paradigm

jasoncrawfordMay 4, 2023, 10:11 PM

122 points

41 votes

Overall karma indicates overall quality.

29 comments13 min readLW link

(rootsofprogress.org)

Recursive Middle Manager Hell: AI Edition

VojtaKovarikMay 4, 2023, 8:08 PM

31 points

12 votes

Overall karma indicates overall quality.

11 comments2 min readLW link

AI risk/reward: A simple model

Nathan YoungMay 4, 2023, 7:25 PM

3 points

1 vote

Overall karma indicates overall quality.

0 comments7 min readLW link

Google “We Have No Moat, And Neither Does OpenAI”

Chris_LeongMay 4, 2023, 6:23 PM

61 points

31 votes

Overall karma indicates overall quality.

28 comments1 min readLW link

(www.semianalysis.com)

Trying to measure AI deception capabilities using temporary simulation fine-tuning

alenoachMay 4, 2023, 5:59 PM

4 points

3 votes

Overall karma indicates overall quality.

0 comments7 min readLW link

[Linkpost]Transformer-Based LM Surprisal Predicts Human Reading Times Best with About Two Billion Training Tokens

Curtis HuebnerMay 4, 2023, 5:16 PM

10 points

6 votes

Overall karma indicates overall quality.

1 comment1 min readLW link

(arxiv.org)

Clarifying and predicting AGI

Richard_NgoMay 4, 2023, 3:55 PM

142 points

66 votes

Overall karma indicates overall quality.

45 comments4 min readLW link

[Crosspost] AI X-risk in the News: How Effective are Recent Media Items and How is Awareness Changing? Our New Survey Results.

otto.bartenMay 4, 2023, 2:09 PM

5 points

3 votes

Overall karma indicates overall quality.

0 comments9 min readLW link

(forum.effectivealtruism.org)

AI #10: Code Interpreter and Geoff Hinton

ZviMay 4, 2023, 2:00 PM

80 points

37 votes

Overall karma indicates overall quality.

7 comments78 min readLW link

(thezvi.wordpress.com)

Advice for interacting with busy people

Severin T. SeehrichMay 4, 2023, 1:31 PM

68 points

41 votes

Overall karma indicates overall quality.

4 comments4 min readLW link

We don’t need AGI for an amazing future

Karl von WendtMay 4, 2023, 12:10 PM

20 points

29 votes

Overall karma indicates overall quality.

32 comments5 min readLW link

Has the Symbol Grounding Problem just gone away?

RussellThorMay 4, 2023, 7:46 AM

12 points

7 votes

Overall karma indicates overall quality.

3 comments1 min readLW link

Opinion merging for AI control

David JohnstonMay 4, 2023, 2:43 AM

6 points

2 votes

Overall karma indicates overall quality.

0 comments11 min readLW link

Understanding why illusionism does not deny the existence of qualia

Mergimio H. DoefevmilMay 4, 2023, 2:13 AM

0 points

8 votes

Overall karma indicates overall quality.

17 comments1 min readLW link

[New] Rejected Content Section

Ruby and Raemon

May 4, 2023, 1:43 AM

65 points

25 votes

Overall karma indicates overall quality.

21 comments5 min readLW link

How MATS addresses “mass movement building” concerns

Ryan KiddMay 4, 2023, 12:55 AM

63 points

39 votes

Overall karma indicates overall quality.

9 comments3 min readLW link

Moving VPS Again

jefftkMay 4, 2023, 12:30 AM

9 points

1 vote

Overall karma indicates overall quality.

2 comments1 min readLW link

(www.jefftk.com)

Prizes for matrix completion problems

paulfchristianoMay 3, 2023, 11:30 PM

164 points

72 votes

Overall karma indicates overall quality.

52 comments1 min readLW link

(www.alignment.org)

Alignment Research @ EleutherAI

Curtis HuebnerMay 3, 2023, 10:45 PM

40 points

22 votes

Overall karma indicates overall quality.

1 comment3 min readLW link

(blog.eleuther.ai)

«Boundaries/Membranes» and AI safety compilation

Chris LakinMay 3, 2023, 9:41 PM

56 points

20 votes

Overall karma indicates overall quality.

17 comments8 min readLW link

[Question] What constraints does deep learning place on alignment plans?

Garrett BakerMay 3, 2023, 8:40 PM

9 points

4 votes

Overall karma indicates overall quality.

0 comments1 min readLW link

Keyboard shortcuts

Keys shown in yellow (e.g., ]) are accesskeys, and require a browser-specific modifier key (or keys).

Keys shown in grey (e.g., ?) do not require any modifier keys.

General
? Show keyboard shortcuts
Esc Hide keyboard shortcuts

Site navigation
h Go to Home (a.k.a. “Frontpage”) view
f Go to Featured (a.k.a. “Curated”) view
a Go to All (a.k.a. “Community”) view
m Go to Meta view
v Go to Tags view
c Go to Recent Comments view
r Go to Archive view
q Go to Sequences view
t Go to About page
u Go to User or Login page
o Go to Inbox page

Page navigation
, Jump up to top of page
. Jump down to bottom of page
/ Jump to top of comments section
s Search

Page actions
n New post or comment
e Edit current post

Post/comment list views
. Focus next entry in list
, Focus previous entry in list
; Cycle between links in focused entry
Enter Go to currently focused entry
Esc Unfocus currently focused entry
] Go to next page
[ Go to previous page
\ Go to first page
e Edit currently focused post

Editor
k Bold text
i Italic text
l Insert hyperlink
q Blockquote text

Appearance
= Increase text size
- Decrease text size
0 Reset to default text size
′ Cycle through content width settings
1 Switch to default theme [A]
2 Switch to dark theme [B]
3 Switch to grey theme [C]
4 Switch to ultramodern theme [D]
5 Switch to simple theme [E]
6 Switch to brutalist theme [F]
7 Switch to ReadTheSequences theme [G]
8 Switch to classic Less Wrong theme [H]
9 Switch to modern Less Wrong theme [I]
; Open theme tweaker
Enter Save changes and close theme tweaker
Esc Close theme tweaker (without saving)

Slide shows
l Start/resume slideshow
Esc Exit slideshow
→↓ Next slide
←↑ Previous slide
Space Reset slide zoom

Miscellaneous
x Switch to next view on user page
z Switch to previous view on user page
` Toggle compact comment list view
g Toggle anti-kibitzer