Learning zero, and what SLT gets wrong about it

Dmitry Vaintrob29 Apr 2026 6:41 UTC

11 points

1 comment13 min readLW link

Are LLMs not getting better?

kqr29 Apr 2026 6:27 UTC

16 points

2 comments2 min readLW link

llm assistant personas seem increasingly incoherent (some subjective observations)

nostalgebraist29 Apr 2026 3:53 UTC

60 points

4 comments9 min readLW link

The AI x-risk lawsuit waiting to happen

David Scott Krueger29 Apr 2026 3:50 UTC

13 points

0 comments2 min readLW link

(therealartificialintelligence.substack.com)

Not a Paper: “Frontier Lab CEOs are Capable of In-Context Scheming”

LawrenceC29 Apr 2026 3:00 UTC

74 points

2 comments7 min readLW link

Notes on Transformer Consciousness

slavachalnev29 Apr 2026 0:00 UTC

23 points

1 comment2 min readLW link

SecureMaxx: A Lightweight Sequence Screening Tool for Agents

Austin Morrissey28 Apr 2026 23:47 UTC

3 points

0 comments8 min readLW link

Will whole brain emulation matter for the AI transition?

djbinder28 Apr 2026 23:04 UTC

19 points

0 comments41 min readLW link

(defensesindepth.bio)

Causal inference diary: skiing causes snow

Gretta Duleba28 Apr 2026 22:21 UTC

15 points

1 comment8 min readLW link

Is AI welfare work puntable?

Oscar28 Apr 2026 21:17 UTC

11 points

2 comments7 min readLW link

The Problem in the “Nerd Sniping” xkcd Comic

peralice28 Apr 2026 20:40 UTC

50 points

3 comments12 min readLW link

Comment on “Forecasting is Way Overrated, and We Should Stop Funding It”

Josh Rosenberg28 Apr 2026 20:16 UTC

22 points

0 comments9 min readLW link

Introspection Adapters: Training LLMs to Report Their Learned Behaviors

keshavs, RowanWang, abhayesian, Sam Marks and SoerenMind

28 Apr 2026 19:02 UTC

21 points

0 comments12 min readLW link

(alignment.anthropic.com)

Recursive forecasting: Eliciting long-term forecasts from myopic fitness-seekers

Jozdien and Alex Mallen

28 Apr 2026 18:00 UTC

55 points

2 comments7 min readLW link

Nobody ever checked

Cameron Berg28 Apr 2026 17:15 UTC

24 points

15 comments8 min readLW link

(camberg.substack.com)

Monday AI Radar #23

Against Moloch28 Apr 2026 17:12 UTC

4 points

0 comments6 min readLW link

(againstmoloch.com)

An Alignment Journal: Adaptation to AI

JessRiedel, Dan MacKinlay, Yonatan Cale and david reinstein

28 Apr 2026 17:04 UTC

22 points

1 comment13 min readLW link

(blog.alignmentjournal.org)

Frontier Coding Agents Can Now Implement an AlphaZero Self-Play Machine Learning Pipeline For Connect Four That Performs Comparably to an External Solver

Baybar, jsherwood and Benjamin Kaplan

28 Apr 2026 16:50 UTC

25 points

0 comments36 min readLW link

Takes from two months as an aspiring LLM naturalist

AnnaSalamon28 Apr 2026 16:14 UTC

95 points

17 comments8 min readLW link

SAEBER: Sparse Autoencoders for Biological Entity Risk

michaelwaves28 Apr 2026 14:43 UTC

8 points

0 comments8 min readLW link