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

All Jan Feb Mar Apr May Jun Jul Aug SepOctNov Dec

All 1 2 3 4 5 6 7 8 9 10 11 121314 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

A strange twist on the road to AGI

cveres12 Oct 2022 23:27 UTC

−8 points

0 comments1 min readLW link

Help out Redwood Research’s interpretability team by finding heuristics implemented by GPT-2 small

Haoxing Du and Buck

12 Oct 2022 21:25 UTC

50 points

11 comments4 min readLW link

Towards a comprehensive study of potential psychological causes of the ordinary range of variation of affective gender identity in males

tailcalled12 Oct 2022 21:10 UTC

52 points

4 comments37 min readLW link

Six (and a half) intuitions for KL divergence

CallumMcDougall12 Oct 2022 21:07 UTC

154 points

25 comments10 min readLW link 1 review

(www.perfectlynormal.co.uk)

[MLSN #6]: Transparency survey, provable robustness, ML models that predict the future

Dan H12 Oct 2022 20:56 UTC

27 points

0 comments6 min readLW link

[Question] Previous Work on Recreating Neural Network Input from Intermediate Layer Activations

bglass12 Oct 2022 19:28 UTC

1 point

3 comments1 min readLW link

Be more effective by learning important practical knowledge using flashcards

Stenemo12 Oct 2022 18:05 UTC

5 points

2 comments1 min readLW link

Article Review: Google’s AlphaTensor

Robert_AIZI12 Oct 2022 18:04 UTC

8 points

4 comments10 min readLW link

Alignment 201 curriculum

Richard_Ngo12 Oct 2022 18:03 UTC

102 points

3 comments1 min readLW link

(www.agisafetyfundamentals.com)

Progress links and tweets, 2022-10-12

jasoncrawford12 Oct 2022 16:59 UTC

8 points

0 comments1 min readLW link

(rootsofprogress.org)

Building a transformer from scratch—AI safety up-skilling challenge

Marius Hobbhahn12 Oct 2022 15:40 UTC

42 points

1 comment5 min readLW link

some simulation hypotheses

Tamsin Leake12 Oct 2022 13:34 UTC

13 points

3 comments5 min readLW link

(carado.moe)

Instrumental convergence in single-agent systems

Edouard Harris and simonsdsuo

12 Oct 2022 12:24 UTC

31 points

4 comments8 min readLW link

(www.gladstone.ai)

Singapore—Small casual dinner in Chinatown #5

Joe Rocca12 Oct 2022 8:59 UTC

3 points

1 comment1 min readLW link

A game of mattering

KatjaGrace12 Oct 2022 8:50 UTC

28 points

2 comments5 min readLW link

(worldspiritsockpuppet.com)

Calibration of a thousand predictions

KatjaGrace12 Oct 2022 8:50 UTC

57 points

7 comments5 min readLW link

(worldspiritsockpuppet.com)

My argument against AGI

cveres12 Oct 2022 6:33 UTC

7 points

5 comments1 min readLW link

Actually, All Nuclear Famine Papers are Bunk

Lao Mein12 Oct 2022 5:58 UTC

113 points

37 comments2 min readLW link 1 review

Contingency is not arbitrary

Gordon Seidoh Worley12 Oct 2022 4:35 UTC

13 points

0 comments3 min readLW link

That one apocalyptic nuclear famine paper is bunk

Lao Mein12 Oct 2022 3:33 UTC

110 points

10 comments1 min readLW link

AstralCodexTen and Rationality Meetup Organisers’ Retreat Asia Pacific region

Elo and Harold

12 Oct 2022 3:20 UTC

14 points

4 comments2 min readLW link

Abbots Bromley Horn Dance History

jefftk12 Oct 2022 2:10 UTC

11 points

0 comments2 min readLW link

(www.jefftk.com)

Power-Seeking AI and Existential Risk

Antonio Franca11 Oct 2022 22:50 UTC

6 points

0 comments9 min readLW link

From technocracy to the counterculture

jasoncrawford11 Oct 2022 19:37 UTC

28 points

1 comment26 min readLW link

(rootsofprogress.org)

Prettified AI Safety Game Cards

abramdemski11 Oct 2022 19:35 UTC

47 points

6 comments1 min readLW link

On the proper piloting of flesh shoots

Mordecai Weynberg11 Oct 2022 18:52 UTC

−4 points

6 comments1 min readLW link

Why I think nuclear war triggered by Russian tactical nukes in Ukraine is unlikely

Dave Orr11 Oct 2022 18:30 UTC

50 points

7 comments3 min readLW link

Anonymous advice: If you want to reduce AI risk, should you take roles that advance AI capabilities?

Benjamin Hilton11 Oct 2022 14:16 UTC

54 points

9 comments1 min readLW link

Misalignment Harms Can Be Caused by Low Intelligence Systems

DialecticEel11 Oct 2022 13:39 UTC

11 points

3 comments1 min readLW link

[Sketch] Validity Criterion for Logical Counterfactuals

DragonGod11 Oct 2022 13:31 UTC

6 points

0 comments4 min readLW link

[Question] How much does the risk of dying from nuclear war differ within and between countries?

amarai11 Oct 2022 11:55 UTC

4 points

7 comments1 min readLW link

Did you enjoy Ramez Naam’s “Nexus” trilogy? Check out this interview on neurotech and the law.

fowlertm11 Oct 2022 11:10 UTC

5 points

0 comments1 min readLW link

What “The Message” Was For Me

Alex Beyman11 Oct 2022 8:08 UTC

−3 points

14 comments4 min readLW link

Updates and Clarifications

SD Marlow11 Oct 2022 5:34 UTC

−5 points

1 comment1 min readLW link

What if human reasoning is anti-inductive?

Q Home11 Oct 2022 5:15 UTC

4 points

2 comments13 min readLW link

Fullness to Indicate Cleanliness

jefftk11 Oct 2022 0:40 UTC

9 points

12 comments1 min readLW link

(www.jefftk.com)

[Question] What happened to the annual LW demographic surveys?

ROM11 Oct 2022 0:19 UTC

5 points

2 comments1 min readLW link

EA & LW Forums Weekly Summary (26 Sep − 9 Oct 22′)

Zoe Williams10 Oct 2022 23:58 UTC

13 points

2 comments1 min readLW link

Don’t expect AGI anytime soon

cveres10 Oct 2022 22:38 UTC

−14 points

6 comments1 min readLW link

QAPR 4: Inductive biases

Quintin Pope10 Oct 2022 22:08 UTC

67 points

2 comments18 min readLW link

Apollo

Jarred Filmer10 Oct 2022 21:30 UTC

46 points

0 comments3 min readLW link

[Question] Does biology reliably find the global maximum, or at least get close?

Noosphere8910 Oct 2022 20:55 UTC

24 points

70 comments1 min readLW link

Disentangling inner alignment failures

Erik Jenner10 Oct 2022 18:50 UTC

20 points

5 comments4 min readLW link

ACX meetup [October]

sallatik10 Oct 2022 17:23 UTC

1 point

0 comments1 min readLW link

Natural Categories Update

Logan Zoellner10 Oct 2022 15:19 UTC

33 points

6 comments2 min readLW link

When reporting AI timelines, be clear who you’re deferring to

Sam Clarke10 Oct 2022 14:24 UTC

38 points

6 comments1 min readLW link

Why Balsa Research is Worthwhile

Zvi10 Oct 2022 13:50 UTC

87 points

12 comments8 min readLW link

(thezvi.wordpress.com)

Lessons learned from talking to >100 academics about AI safety

Marius Hobbhahn10 Oct 2022 13:16 UTC

214 points

17 comments12 min readLW link 1 review

We can do better than argmax

Jan_Kulveit10 Oct 2022 10:32 UTC

48 points

4 comments1 min readLW link

Vegetarianism and depression

Maggy10 Oct 2022 9:11 UTC

2 points

2 comments1 min readLW link