The Rogue Agent Explosion Will Be Mostly Invisible

Steven McCulloch19 Aug 2026 18:46 UTC

18 points

1 comment16 min readLW link

RL creates split personas

Jan Betley19 Aug 2026 18:23 UTC

47 points

0 comments4 min readLW link

A failed solution to open-source game theory

Richard Willis19 Aug 2026 16:11 UTC

14 points

2 comments4 min readLW link

Inside the mind of a fair player cooperating

transhumanist_atom_understander19 Aug 2026 14:53 UTC

18 points

0 comments3 min readLW link

Some reasons alignment doesn’t generalise well

Lucius Bushnaq19 Aug 2026 14:18 UTC

54 points

0 comments9 min readLW link

The Control Paradox

Ephraiem Sarabamoun19 Aug 2026 12:39 UTC

5 points

0 comments2 min readLW link

Concerns About Personas, Multi-Agent Alignment, and Role Theory

Davidmanheim19 Aug 2026 12:03 UTC

18 points

0 comments8 min readLW link

Reading List on Wise AI

Chris_Leong19 Aug 2026 8:41 UTC

19 points

0 comments1 min readLW link

What AI scores (while we can still keep score)

dan.parshall19 Aug 2026 1:40 UTC

11 points

2 comments3 min readLW link

Funding Formal Methods for the Cyberpocalypse

Max von Hippel18 Aug 2026 16:45 UTC

21 points

6 comments9 min readLW link

Policy career planning in the age of imminent superintelligence

Peter Wildeford18 Aug 2026 14:33 UTC

60 points

2 comments6 min readLW link

Automatic Programming Should Be More Like SQL

Adam Chlipala18 Aug 2026 12:29 UTC

7 points

0 comments12 min readLW link

AI Security is Harm Reduction

Quinn18 Aug 2026 12:28 UTC

39 points

1 comment1 min readLW link

Price recursion is the rational theory of reward

Abhimanyu Pallavi Sudhir18 Aug 2026 3:22 UTC

6 points

0 comments9 min readLW link

Misaligned Incentives in Pause Scenarios

Michael Soareverix and Antra Tessera

18 Aug 2026 1:55 UTC

39 points

10 comments17 min readLW link

For Claude, capability and dispreferring CDT are the ~same thing. Much more so than for GPT.

Chi Nguyen and Emery Cooper

18 Aug 2026 1:25 UTC

47 points

11 comments1 min readLW link

You Can’t Iterate to Trustworthy AI Code Without Understanding

ronbodkin17 Aug 2026 23:00 UTC

8 points

0 comments10 min readLW link

What gives you away: how LLMs form opinions of you

Cat McGee17 Aug 2026 20:59 UTC

37 points

7 comments6 min readLW link

Evaluating Chain-of-Thought Monitorability is Still an Open Problem: Comments on OpenAI’s Monitorability Evals

Connor Dilgren and Sarah Wiegreffe

17 Aug 2026 19:45 UTC

17 points

1 comment18 min readLW link

Weird Re-Tokenization, Symmetries and Compression: Research Agenda

Xenomirant and Sami Wolf

17 Aug 2026 19:43 UTC

19 points

14 comments16 min readLW link