8 Jan 2026 22:22 UTC

64 points

4 comments18 min readLW link

(post-agi.org)

The Hunger Strike To Stop The AI Race

Michaël Trazzi8 Jan 2026 21:05 UTC

37 points

0 comments1 min readLW link

(www.youtube.com)

Skepticism about Introspection in LLMs

derek shiller8 Jan 2026 20:07 UTC

11 points

7 comments36 min readLW link

On rationality skills

dominicq8 Jan 2026 19:23 UTC

3 points

3 comments1 min readLW link

(sundaystopwatch.eu)

Self-Help Tactics That Are Working For Me

sarahconstantin8 Jan 2026 18:00 UTC

54 points

2 comments11 min readLW link

(sarahconstantin.substack.com)

Distinguishing Sensory Qualia Using Neural Structure

Shiva's Right Foot8 Jan 2026 16:45 UTC

4 points

0 comments8 min readLW link

Why LLMs Aren’t Scientists Yet.

Dhruv Trehan8 Jan 2026 16:06 UTC

39 points

3 comments5 min readLW link

(arxiv.org)

Can We Make AI Alignment Framing Less Wrong?

Anurag 8 Jan 2026 15:20 UTC

3 points

0 comments4 min readLW link

AI #150: While Claude Codes

Zvi8 Jan 2026 15:00 UTC

42 points

3 comments20 min readLW link

(thezvi.wordpress.com)

Saying What You Want

omegastick8 Jan 2026 14:12 UTC

17 points

0 comments3 min readLW link

(dumbideas.xyz)

Small Steps Towards Proving Stochastic → Deterministic Natural Latent

Alfred Harwood and Jeremy Gillen

8 Jan 2026 12:27 UTC

57 points

4 comments10 min readLW link

Using Anki to memorise the names of the MATS 9 cohort

beyarkay (Boyd Kane)8 Jan 2026 4:41 UTC

5 points

0 comments3 min readLW link

(boydkane.com)

Rents Are High, But Not Skyrocketing

jefftk8 Jan 2026 2:40 UTC

29 points

1 comment1 min readLW link

(www.jefftk.com)

The AI Infrastructure Security Shortlist

Abbey Chaver8 Jan 2026 2:26 UTC

21 points

0 comments6 min readLW link

HIA and X-risk part 2: Why it hurts

TsviBT8 Jan 2026 1:19 UTC

63 points

13 comments21 min readLW link

Beliefs and position going into 2026

RussellThor8 Jan 2026 1:11 UTC

5 points

0 comments5 min readLW link

Lumina Probiotic worked for me!

Eye You8 Jan 2026 0:34 UTC

46 points

3 comments2 min readLW link

Taiwan Trip Report

nomagicpill7 Jan 2026 23:40 UTC

11 points

0 comments9 min readLW link

(nomagicpill.substack.com)

Public intellectuals need to say what they actually believe

Aaron Bergman7 Jan 2026 21:22 UTC

79 points

12 comments14 min readLW link

(www.aaronbergman.net)

Two Aspects of Situational Awareness: World Modelling & Indexical Information

David Scott Krueger7 Jan 2026 20:24 UTC

40 points

7 comments2 min readLW link

Advancements In Self-Driving Cars

Zvi7 Jan 2026 19:50 UTC

30 points

2 comments17 min readLW link

(thezvi.wordpress.com)

Two ways non-U.S. folks can contribute to AI going well

Joe Rogero7 Jan 2026 19:37 UTC

21 points

1 comment2 min readLW link

(subatomicarticles.com)

Everything is Political Now, or, A Review of “Fraggle Rock: Back to the Rock”

Gordon Seidoh Worley7 Jan 2026 17:00 UTC

13 points

0 comments8 min readLW link

(www.uncertainupdates.com)

FirstPrinciples Talks: Quantum machines learning quantum

Carly Turini7 Jan 2026 16:44 UTC

3 points

0 comments1 min readLW link

Does mindfulness meditation lead to awakening?

Vadim Golub7 Jan 2026 14:08 UTC

16 points

0 comments4 min readLW link

An interactive toy model for exploring AI’s effect on the labour market

CharlesD7 Jan 2026 12:57 UTC

12 points

0 comments7 min readLW link

OpenForecaster: How to train language models for open-ended forecasting?

nikhilchandak, shash42 and bayesian_kitten

7 Jan 2026 11:03 UTC

10 points

1 comment7 min readLW link

ML research directions for preventing catastrophic data poisoning

Tom Davidson7 Jan 2026 10:16 UTC

35 points

1 comment10 min readLW link

(newsletter.forethought.org)

A Loser’s Reflections

L.M.Sherlock7 Jan 2026 7:15 UTC

9 points

12 comments18 min readLW link

(lmsherlock.substack.com)

Algorithmic Dating

denzit7 Jan 2026 2:39 UTC

−2 points

0 comments3 min readLW link

(denzit.substack.com)

Simple summary of AI Safety laws

Joseph Miller, Espedair Street and PauseAI UK

7 Jan 2026 1:51 UTC

46 points

4 comments3 min readLW link

Results: A self-randomized study of the impacts of glycine on sleep (Science is still hard)

thedissonance.net6 Jan 2026 20:54 UTC

6 points

1 comment3 min readLW link

(thedissonance.net)

My 2003 Post on the Evolutionary Argument for AI Misalignment

Wei Dai6 Jan 2026 20:45 UTC

37 points

7 comments2 min readLW link

Mainstream approach for alignment evals is a dead end

Igor Ivanov6 Jan 2026 19:52 UTC

60 points

9 comments5 min readLW link

Fertility Roundup #6: The Art of More Dakka

Zvi6 Jan 2026 19:50 UTC

32 points

5 comments26 min readLW link

(thezvi.wordpress.com)

On Owning Galaxies

Simon Lermen6 Jan 2026 18:16 UTC

154 points

62 comments3 min readLW link

(simonlermen.substack.com)

How hard is it to inoculate against misalignment generalization?

Jozdien6 Jan 2026 17:30 UTC

46 points

4 comments14 min readLW link

How AI Is Learning to Think in Secret

Nicholas Andresen6 Jan 2026 16:31 UTC

382 points

32 comments18 min readLW link

(nickandresen.substack.com)

Should you be posting on the open internet

zef6 Jan 2026 15:50 UTC

22 points

9 comments2 min readLW link

Catching misreporting about ML hardware use by turning noise into signal—Part II

Naci Cankaya6 Jan 2026 12:38 UTC

8 points

0 comments1 min readLW link

(nacicankaya.substack.com)

Meditations on Moloch in the AI Rat Race

Alexander Müller6 Jan 2026 9:46 UTC

11 points

1 comment6 min readLW link

[Question] Is anyone doing a real-world test of agentic misalignment?

Jamie Milton Freestone6 Jan 2026 7:45 UTC

2 points

1 comment1 min readLW link

Do we need sparsity afterall?

Giuseppe Birardi6 Jan 2026 6:06 UTC

20 points

5 comments29 min readLW link

Exploring Reinforcement Learning Effects on Chain-of-Thought Legibility

Julian H, RohanS, Baram Sosis, vedant-badoni and The-Turtle

6 Jan 2026 3:04 UTC

41 points

3 comments21 min readLW link

The Evolution Argument Sucks

peralice6 Jan 2026 2:32 UTC

30 points

6 comments8 min readLW link

Festival Stats 2025

jefftk6 Jan 2026 1:40 UTC

10 points

1 comment1 min readLW link

(www.jefftk.com)

Oversight Assistants: Turning Compute into Understanding

jsteinhardt6 Jan 2026 0:50 UTC

85 points

7 comments9 min readLW link

(bounded-regret.ghost.io)

Aether is hiring technical AI safety researchers

Rauno Arike, RohanS and Shubhorup Biswas

5 Jan 2026 22:27 UTC

22 points

0 comments2 min readLW link

[Question] Continual Learning Achieved?

PeterMcCluskey5 Jan 2026 22:22 UTC

−7 points

11 comments1 min readLW link

AGI will not be one specific system, it’ll be the unity of all systems

henophilia5 Jan 2026 18:21 UTC

−4 points

0 comments11 min readLW link