All 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 20252026

AllJanFeb Mar Apr May Jun

All 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 222324 25 26 27 28 29 30 31

Like night and day: Light glasses and dark therapy can treat non-24 (and SAD)

JennaS22 Jan 2026 23:23 UTC

30 points

1 comment9 min readLW link

Does Pentagon Pizza Theory Work?

rba22 Jan 2026 19:24 UTC

140 points

11 comments5 min readLW link

(goflaw.substack.com)

The phases of an AI takeover

sjadler22 Jan 2026 19:09 UTC

12 points

1 comment9 min readLW link

(stevenadler.substack.com)

Will we get automated alignment research before an AI Takeoff?

Jan Wehner22 Jan 2026 17:46 UTC

33 points

2 comments11 min readLW link

[Question] How Could I Have Learned That Faster?

Dom Polsinelli22 Jan 2026 17:35 UTC

9 points

4 comments2 min readLW link

AI can suddenly become dangerous despite gradual progress

Simon Lermen22 Jan 2026 16:47 UTC

15 points

0 comments4 min readLW link

(simonlermen.substack.com)

Releasing TakeOverBench.com: a benchmark, for AI takeover

otto.barten22 Jan 2026 16:34 UTC

16 points

5 comments1 min readLW link

AI #152: Brought To You By The Torment Nexus

Zvi22 Jan 2026 14:40 UTC

35 points

5 comments56 min readLW link

(thezvi.wordpress.com)

Resisting Reality

robertzk22 Jan 2026 13:50 UTC

26 points

3 comments6 min readLW link

Experiments on Reward Hacking Monitorability in Language Models

Monketo22 Jan 2026 2:42 UTC

9 points

0 comments8 min readLW link

Neural chameleons can(’t) hide from activation oracles

ceselder22 Jan 2026 1:47 UTC

55 points

5 comments3 min readLW link

Dedicated continuous supervision of AI companies

Michael Bennett22 Jan 2026 1:47 UTC

8 points

0 comments15 min readLW link

Uncovering Unfaithful CoT in Deceptive Models

Agastya Agrawal22 Jan 2026 1:46 UTC

12 points

2 comments3 min readLW link

Claude’s Constitution is an excellent guide for humans, too

Eye You22 Jan 2026 1:26 UTC

27 points

0 comments5 min readLW link

The first type of transformative AI?

Lizka21 Jan 2026 23:47 UTC

19 points

0 comments1 min readLW link

(www.forethought.org)

How (and why) to read Drexler on AI

owencb21 Jan 2026 23:25 UTC

55 points

12 comments6 min readLW link

(strangecities.substack.com)

Finding Yourself in Others

1a3orn21 Jan 2026 23:22 UTC

51 points

1 comment4 min readLW link

AI Risks Slip Out of Mind

MarkelKori21 Jan 2026 22:30 UTC

5 points

1 comment1 min readLW link

When should we train against a scheming monitor?

Mary Phuong21 Jan 2026 20:48 UTC

24 points

4 comments5 min readLW link

Claude Codes #3

Zvi21 Jan 2026 19:50 UTC

47 points

5 comments15 min readLW link

(thezvi.wordpress.com)

Claude’s new constitution

Zac Hatfield-Dodds and Drake Thomas

21 Jan 2026 19:37 UTC

176 points

47 comments6 min readLW link

(www.anthropic.com)

Crimes of the Future, Solutions of the Past

evrim21 Jan 2026 19:20 UTC

18 points

1 comment4 min readLW link

On visions of a “good future” for humanity in a world with artificial superintelligence

Jakub Growiec21 Jan 2026 18:27 UTC

1 point

0 comments30 min readLW link

The case for AGI safety products

Marius Hobbhahn21 Jan 2026 17:23 UTC

68 points

7 comments12 min readLW link

Updating in the Opposite Direction from Evidence

Dom Polsinelli21 Jan 2026 16:08 UTC

1 point

0 comments3 min readLW link

(dompols.substack.com)

Vibing with Claude, January 2026 Edition

Gordon Seidoh Worley21 Jan 2026 16:00 UTC

26 points

2 comments4 min readLW link

(www.uncertainupdates.com)

AI Needs People (So, It Won’t Be Like Terminator Movie)

Victor Porton21 Jan 2026 14:42 UTC

−23 points

0 comments2 min readLW link

Kredit Grant

kian21 Jan 2026 0:56 UTC

5 points

5 comments1 min readLW link

Money Can’t Buy the Smile on a Child’s Face As They Look at A Beautiful Sunset… but it also can’t buy a malaria free world: my current understanding of how Effective Altruism has failed

Hazard20 Jan 2026 23:28 UTC

70 points

17 comments6 min readLW link

(naturalhazard.xyz)

ACX Atlanta February Meetup

Steve French20 Jan 2026 22:30 UTC

2 points

0 comments1 min readLW link

So Long Sucker: AI Deception, “Alliance Banks,” and Institutional Lying

fernando yt20 Jan 2026 22:29 UTC

47 points

5 comments2 min readLW link

No instrumental convergence without AI psychology

TurnTrout20 Jan 2026 22:16 UTC

68 points

7 comments6 min readLW link

(turntrout.com)

MLSN #18: Adversarial Diffusion, Activation Oracles, Weird Generalization

Alice Blair and Dan H

20 Jan 2026 17:03 UTC

14 points

3 comments5 min readLW link

Against “If Anyone Builds It Everyone Dies”

Bentham's Bulldog20 Jan 2026 16:49 UTC

6 points

9 comments22 min readLW link

ChatGPT Self Portrait

Zvi20 Jan 2026 16:30 UTC

61 points

10 comments3 min readLW link

(thezvi.wordpress.com)

Deep learning as program synthesis

Zach Furman20 Jan 2026 15:35 UTC

150 points

33 comments41 min readLW link

The Total Solar Eclipse of 2238 and GPT-5.2 Pro

spookyuser20 Jan 2026 14:27 UTC

5 points

2 comments5 min readLW link

Free New Year’s Workshop — Elemental Lenses for 2026

teebarnett20 Jan 2026 10:47 UTC

1 point

0 comments1 min readLW link

Why I Transitioned: A Response

quinoa marisa20 Jan 2026 2:06 UTC

156 points

47 comments10 min readLW link

Appendix: Contra Fiora on Contra

quinoa marisa20 Jan 2026 1:53 UTC

14 points

0 comments7 min readLW link

A Criteron for Deception

Mariven20 Jan 2026 1:25 UTC

8 points

2 comments4 min readLW link

(mariven.substack.com)

Evidence that would update me towards a software-only fast takeoff

Anders Cairns Woodruff20 Jan 2026 0:58 UTC

15 points

4 comments4 min readLW link

There may be low hanging fruit for a weak nootropic

Dom Polsinelli20 Jan 2026 0:51 UTC

31 points

6 comments4 min readLW link

(open.substack.com)

Everybody Wants to Rule the Future—Is Longtermism’s Mandate of Heaven by Arithmetic Justified?

E.G. Blee-Goldman19 Jan 2026 23:31 UTC

6 points

10 comments9 min readLW link

What can Kickstarter teach us about goal completion?

Elijah19 Jan 2026 22:03 UTC

13 points

0 comments4 min readLW link

All (Non-Trivial) Decisions Are Undecidable

(M)ason19 Jan 2026 21:51 UTC

−9 points

1 comment1 min readLW link

Pretraining on Aligned AI Data Dramatically Reduces Misalignment—Even After Post-Training

RogerDearnaley19 Jan 2026 21:24 UTC

106 points

12 comments11 min readLW link

(arxiv.org)

Medical Roundup #6

Zvi19 Jan 2026 21:20 UTC

31 points

2 comments11 min readLW link

(thezvi.wordpress.com)

Could LLM alignment research reduce x-risk if the first takeover-capable AI is not an LLM?

Tim Hua19 Jan 2026 18:09 UTC

25 points

2 comments6 min readLW link

AGI both does and doesn’t have an infinite time horizon

Sean Herrington19 Jan 2026 16:57 UTC

15 points

0 comments4 min readLW link