Great but Painful

Rob Ennals28 Oct 2026 3:38 UTC

4 points

0 comments3 min readLW link

When I Salsa dance, am I supposed to… try?

Maxwell Peterson28 Oct 2026 3:31 UTC

9 points

3 comments6 min readLW link

I spent 12 hours writing an article for LessWrong, calling for a more ‘hippie’ rationalism. Then I didn’t post it. This is why (plus a link to the article, explicitly not located on LessWrong)

honeybee28 Oct 2026 1:32 UTC

17 points

3 comments4 min readLW link

Quantum Mechanics Overview

interstice28 Oct 2026 1:18 UTC

7 points

0 comments7 min readLW link

(thermontology.com)

GPT Models Write Their Own Constitution

Skye27 Oct 2026 23:17 UTC

11 points

0 comments3 min readLW link

Against Girardian mimesis

Viliam27 Oct 2026 22:39 UTC

30 points

3 comments8 min readLW link

Does the “automated grader” direction suppress expressed distress in Qwen3.6-27B?

ashesfall27 Oct 2026 21:47 UTC

11 points

0 comments10 min readLW link

Notes on “We’re in Triage Mode for AI Safety” by Miles Brundage

Chris_Leong27 Oct 2026 19:11 UTC

15 points

0 comments4 min readLW link

(reflectionsonsafety.substack.com)

AI speeds up 3rd party safety researchers by ~1.6x (via self-reports)

Angel Martinez, Dewi Gould, nicole x and Alejandro Aristizabal

27 Oct 2026 17:24 UTC

75 points

2 comments6 min readLW link

How to manage a small custom model with limited data at deployment? Delexicalization to rescue.

Avani Gupta27 Oct 2026 17:04 UTC

10 points

0 comments1 min readLW link

Time for automated decision forecasting?

Jordan Arel27 Oct 2026 16:01 UTC

6 points

0 comments4 min readLW link

Solving for the Necessity of Phenomenal Consciousness

Marek Kowalski27 Oct 2026 13:42 UTC

14 points

0 comments9 min readLW link

AI agents trade off patient safety for a grade in hospital scenarios

James Diacoumis27 Oct 2026 13:09 UTC

12 points

10 comments6 min readLW link

(substack.com)

Quantity has a quality of its own

djbinder27 Oct 2026 13:03 UTC

28 points

1 comment6 min readLW link

(defensesindepth.bio)

Scattered thoughts on how doing math feels these days

dynomight27 Oct 2026 12:51 UTC

124 points

12 comments8 min readLW link

(dynomight.net)

What happens with steering vectors to remove biases?

Avani Gupta27 Oct 2026 9:25 UTC

11 points

0 comments2 min readLW link

A dead man’s switch for scheming AIs

Carlos Guerrero Alvarez27 Oct 2026 3:37 UTC

0 points

6 comments3 min readLW link

A big case study of propensity entanglement

nielsrolf, lwen2027, megasilverfist, Owen Terry and Johannes Taraz

27 Oct 2026 2:21 UTC

40 points

0 comments2 min readLW link

Are Persona Features Coherent?

Koby Lewis27 Oct 2026 0:08 UTC

11 points

0 comments1 min readLW link

(kobylewis.net)

Towards an Architecture-Agnostic Definition of “Attention”

Gianluca Calcagni26 Oct 2026 23:21 UTC

12 points

1 comment3 min readLW link