My Cognitive Architecture: A Self-Observational Map

Naj Ami-Nave25 Mar 2026 2:23 UTC

1 point

0 comments8 min readLW link

Is Gemini 3 Scheming in the Wild?

Alejandro Wainstock, Agustin_Martinez_Suñe, Iván Arcuschin and Victor Braberman

25 Mar 2026 1:12 UTC

41 points

1 comment17 min readLW link

Book Review: Open Socrates (Part 1)

Zvi24 Mar 2026 22:21 UTC

22 points

2 comments116 min readLW link

(thezvi.wordpress.com)

Book Review: Open Socrates (Part 2)

Zvi24 Mar 2026 22:20 UTC

13 points

1 comment81 min readLW link

(thezvi.wordpress.com)

Agents Can Get Stuck in Self-distrusting Equilibria

Ashe Vazquez Nuñez24 Mar 2026 22:05 UTC

26 points

1 comment12 min readLW link

Latent Introspection (and other open-source introspection papers)

vgel24 Mar 2026 21:23 UTC

79 points

1 comment9 min readLW link

(arxiv.org)

An Informal Definition of Goals for Embedded Agents

Ashe Vazquez Nuñez24 Mar 2026 18:36 UTC

10 points

0 comments1 min readLW link

My cost-effectiveness unit

Zach Stein-Perlman24 Mar 2026 15:30 UTC

42 points

3 comments4 min readLW link

The Fourth World

Linch24 Mar 2026 13:43 UTC

17 points

9 comments6 min readLW link

Safe Recursive Self-Improvement with Verified Compilers

Adam Chlipala24 Mar 2026 13:35 UTC

9 points

0 comments11 min readLW link

Comparing Across Possible Worlds

unruly abstractions24 Mar 2026 10:09 UTC

7 points

3 comments5 min readLW link

The AIXI perspective on AI Safety

Cole Wyeth24 Mar 2026 3:24 UTC

69 points

0 comments6 min readLW link

Information Overdose

0xmadlad24 Mar 2026 2:26 UTC

−2 points

0 comments3 min readLW link

We cannot safely automate value alignment evaluation and research without thinking about delegation and discretion

Maria Federica Martino Lena 24 Mar 2026 2:25 UTC

2 points

0 comments10 min readLW link

Every Major LLM is a 1-Box Smoking Thirder

Olivia Scharfman24 Mar 2026 2:18 UTC

10 points

4 comments10 min readLW link

A ToM-Inspired Agenda for AI Safety Research

Andrés Cotton24 Mar 2026 2:13 UTC

7 points

0 comments5 min readLW link

Poking and Editing the Circuits

unruly abstractions24 Mar 2026 1:15 UTC

2 points

0 comments8 min readLW link

Adopt a debugger’s mindset to solve your recurring life problems

Declan Molony24 Mar 2026 0:22 UTC

21 points

3 comments8 min readLW link

Experimental Evidence for Simulator Theory— Part 2: The Scalers Strike Back

RogerDearnaley23 Mar 2026 22:37 UTC

13 points

0 comments34 min readLW link

Experimental Evidence for Simulator Theory— Part 1: Emergent Misalignment and Weird Generalizations

RogerDearnaley23 Mar 2026 22:37 UTC

19 points

0 comments53 min readLW link