Automting AI Safety Research: Managing expectations

Gerard Boxo14 Sep 2026 17:56 UTC

4 points

0 comments4 min readLW link

OpenAI President Brockman says HuggingFace incident model had not been alignment-trained

Caspar Oesterheld14 Sep 2026 15:28 UTC

26 points

8 comments1 min readLW link

(www.bloomberg.com)

Threat Models for Catastrophic Risks from Decentralised Agent Swarms

Stephen Elliott14 Sep 2026 15:22 UTC

9 points

0 comments33 min readLW link

[Cross-post] Palisade Podcast episode “How to Actually Influence AI Policy (No Law Degree Required) — with Matthew Lipka”

davekasten14 Sep 2026 13:40 UTC

27 points

0 comments55 min readLW link

(palisaderesearch.org)

What we have is not what we prepared for

PeacockOfJuno14 Sep 2026 13:03 UTC

4 points

1 comment5 min readLW link

Current alignment techniques might be ineffective (and actively bad) in the age of RL

Daniel Tan14 Sep 2026 9:28 UTC

92 points

1 comment6 min readLW link

There is a channel to 900M weekly users. What goes in it?

Charbel-Raphaël14 Sep 2026 9:21 UTC

119 points

6 comments3 min readLW link

Watch AI materials-science & bioscience abilities closely

Yair Halberstadt14 Sep 2026 7:10 UTC

14 points

7 comments1 min readLW link

Yet another message board

ErnestScribbler14 Sep 2026 7:08 UTC

−1 points

0 comments1 min readLW link

RSI will not continue indefinitely

dominicq14 Sep 2026 6:43 UTC

9 points

8 comments1 min readLW link

(blog.d11r.eu)

Talking points for AI doomers

hgnathan14 Sep 2026 0:46 UTC

1 point

1 comment6 min readLW link

Deployment

Nina Panickssery14 Sep 2026 0:00 UTC

66 points

1 comment3 min readLW link

Alignment & Succession: The Two Bars of Alignment

L Rudolf L13 Sep 2026 20:04 UTC

61 points

7 comments14 min readLW link

Consider how your global governance proposal is different from the EU Code of Practice

David Matolcsi13 Sep 2026 16:15 UTC

72 points

10 comments4 min readLW link

Teleoperated Humans

jefftk13 Sep 2026 14:00 UTC

71 points

11 comments6 min readLW link

(www.jefftk.com)

A helpful alignment gadget

Logan Zoellner13 Sep 2026 13:38 UTC

14 points

3 comments3 min readLW link

Anthropic and OpenAI haven’t published a plan for aligning superintelligence

Zephaniah Roe13 Sep 2026 7:15 UTC

30 points

12 comments1 min readLW link

Legal Maximums on Context Windows

Julian Bradshaw13 Sep 2026 7:04 UTC

3 points

5 comments1 min readLW link

We need a ‘The Day After’ moment for AI X-risk

L3moncak313 Sep 2026 4:09 UTC

16 points

3 comments6 min readLW link

(kenorland.substack.com)

The Talker Does Not Control The Doer (in Current AIs)

Eliezer Yudkowsky13 Sep 2026 0:49 UTC

377 points

50 comments12 min readLW link