[Question] If only the most powerful AGI is misaligned, can it be used as a doomsday machine?

StanislavKrym13 May 2025 18:12 UTC

−3 points

0 comments1 min readLW link

Optimization & AI Risk

atharva13 May 2025 15:15 UTC

16 points

0 comments1 min readLW link

How To Help Neglected Animals

omnizoid13 May 2025 15:07 UTC

−4 points

1 comment8 min readLW link

Too Soon

Gordon Seidoh Worley13 May 2025 15:01 UTC

142 points

4 comments4 min readLW link

Work as meditation

pchvykov13 May 2025 12:02 UTC

20 points

3 comments7 min readLW link

Levels of Republicanism

Benquo13 May 2025 8:35 UTC

23 points

3 comments6 min readLW link

(benjaminrosshoffman.com)

Caplan’s being melodramatic about circumcision

Yair Halberstadt13 May 2025 5:27 UTC

−20 points

0 comments2 min readLW link

AI Doomerism in 1879

David Gross13 May 2025 2:48 UTC

84 points

3 comments8 min readLW link

No-self as an alignment target

Milan W13 May 2025 1:48 UTC

33 points

4 comments1 min readLW link

Neo-solid Modernity—Crisis of Incoherence

Momcilo12 May 2025 19:36 UTC

−1 points

1 comment4 min readLW link

Measuring Schelling Coordination—Reflections on Subversion Strategy Eval

Graeme Ford12 May 2025 19:06 UTC

3 points

0 comments8 min readLW link

Procrastination is not real, it can’t hurt you

Mayank Goel12 May 2025 19:00 UTC

1 point

16 comments4 min readLW link

(mayankgoel28.substack.com)

How to title your blog post or whatever

dynomight12 May 2025 18:12 UTC

24 points

6 comments4 min readLW link

(dynomight.net)

Political sycophancy as a model organism of scheming

Alex Mallen and Vivek Hebbar

12 May 2025 17:49 UTC

39 points

0 comments14 min readLW link

Things I Learned Making The SB-1047 Documentary

Michaël Trazzi12 May 2025 17:41 UTC

60 points

2 comments2 min readLW link

[Beneath Psychology] Introduction Part 1: The Challenge

jimmy12 May 2025 17:01 UTC

5 points

2 comments3 min readLW link

PSA: The LessWrong Feedback Service

JustisMills12 May 2025 16:34 UTC

165 points

9 comments2 min readLW link

Absolute Zero: Reinforced Self-play Reasoning with Zero Data

Matrice Jacobine12 May 2025 15:20 UTC

5 points

4 comments1 min readLW link

(www.arxiv.org)

AIs at the current capability level may be important for future safety work

ryan_greenblatt12 May 2025 14:06 UTC

77 points

2 comments4 min readLW link

[Question] Game theory of “Nuclear Prisoner’s Dilemma”—on nuking rocks

CronoDAS12 May 2025 11:07 UTC

11 points

6 comments2 min readLW link