CommentBench: Can Models Match Human Comments on AI Safety Posts?

OscarGilg and Hasan Baig

19 Sep 2026 14:27 UTC

13 points

0 comments8 min readLW link

The AI race is already multipolar

June Jimenez19 Sep 2026 13:51 UTC

14 points

0 comments4 min readLW link

The DT Calyx space probe

DecoRayder19 Sep 2026 9:59 UTC

1 point

0 comments12 min readLW link

Learnings from a week in the wet lab

michaelwaves19 Sep 2026 6:55 UTC

18 points

0 comments4 min readLW link

Gemini had its first breakout: Google claims it is not misalignment?

Young Jae Koh19 Sep 2026 2:34 UTC

9 points

0 comments1 min readLW link

Koopman Theory and Metaethics

bruberu19 Sep 2026 2:28 UTC

10 points

4 comments15 min readLW link

My Current Model of What Happened to Elon Musk

KatSpartz18 Sep 2026 23:15 UTC

10 points

8 comments5 min readLW link

Pretraining data, not verifiability, is why LLMs are especially good at math (and coding)

Steven Byrnes18 Sep 2026 17:08 UTC

106 points

39 comments2 min readLW link

Stopgap Measures to Address Immediate AI Security Threats

Andrea_Miotti and Gabriel Alfour

18 Sep 2026 16:52 UTC

32 points

1 comment12 min readLW link

(blog.controlai.org)

Persuasion Undermining Control: Can AI Talk its Way Out of Human Control?

Josh Levy, myang_ and Kellin Pelrine

18 Sep 2026 16:34 UTC

11 points

0 comments7 min readLW link

(www.far.ai)

The J-Space Debate, Agent Swarms, and Pacing Frontier AI—Digital Minds Newsletter #4

18 Sep 2026 16:09 UTC

9 points

0 comments37 min readLW link

(digitalminds.substack.com)

My Reflections Towards the Path to Greatness

Jv Thunder18 Sep 2026 15:33 UTC

4 points

1 comment5 min readLW link

Announcing Formal Verification at RESI (The Institute for Responsible Superintelligence)

Adam Chlipala18 Sep 2026 14:32 UTC

13 points

0 comments6 min readLW link

Collective Epistemics: Napkin Math on Independent Errors

Jonas Hallgren18 Sep 2026 13:42 UTC

26 points

1 comment7 min readLW link

The Game is Set for a Targeted Memetic Attack on the AI Safety Community

keltan18 Sep 2026 5:43 UTC

112 points

16 comments1 min readLW link

The Horse

Character#273618 Sep 2026 2:52 UTC

58 points

3 comments3 min readLW link

Deep recurrent models are less robustly CoT-monitorable than normal CoT models in a toy setting

Nick Kuhn, evhu and Alek Westover

18 Sep 2026 2:47 UTC

85 points

1 comment11 min readLW link

Machine intelligence and the death of human expression

Girard Dorney18 Sep 2026 2:10 UTC

6 points

0 comments6 min readLW link

(extinctiondesk.substack.com)

The Cost of Utopias (a Dialog)

WillPetillo18 Sep 2026 1:57 UTC

16 points

2 comments17 min readLW link

Two Axes of Alignment: A Framework for Robust Superintelligence Alignment

Arihant Gadgade18 Sep 2026 0:59 UTC

7 points

0 comments4 min readLW link