24 Dec 2024 22:45 UTC

46 points

4 comments91 min readLW link

(thebayesianconspiracy.substack.com)

Acknowledging Background Information with P(Q|I)

JenniferRM24 Dec 2024 18:50 UTC

29 points

8 comments14 min readLW link

Game Theory and Behavioral Economics in The Stock Market

Jaiveer Singh24 Dec 2024 18:15 UTC

1 point

0 comments3 min readLW link

[Question] What are the main arguments against AGI?

Edy Nastase24 Dec 2024 15:49 UTC

1 point

6 comments1 min readLW link

[Question] Recommendations on communities that discuss AI applications in society

Annapurna24 Dec 2024 13:37 UTC

7 points

2 comments1 min readLW link

AIs Will Increasingly Fake Alignment

Zvi24 Dec 2024 13:00 UTC

89 points

0 comments52 min readLW link

(thezvi.wordpress.com)

Apply to the 2025 PIBBSS Summer Research Fellowship

DusanDNesic and Lucas Teixeira

24 Dec 2024 10:25 UTC

15 points

0 comments2 min readLW link

Human-AI Complementarity: A Goal for Amplified Oversight

rishubjain and Sophie Bridgers

24 Dec 2024 9:57 UTC

27 points

4 comments1 min readLW link

(deepmindsafetyresearch.medium.com)

Preliminary Thoughts on Flirting Theory

Alice Blair24 Dec 2024 7:37 UTC

16 points

6 comments3 min readLW link

[Question] Why is neuron count of human brain relevant to AI timelines?

samuelshadrach24 Dec 2024 5:15 UTC

6 points

7 comments1 min readLW link

How Much to Give is a Pragmatic Question

jefftk24 Dec 2024 4:20 UTC

12 points

1 comment2 min readLW link

(www.jefftk.com)

Do you need a better map of your myriad of maps to the territory?

CstineSublime24 Dec 2024 2:00 UTC

11 points

2 comments5 min readLW link

Panology

JenniferRM23 Dec 2024 21:40 UTC

17 points

10 comments5 min readLW link

Aristotle, Aquinas, and the Evolution of Teleology: From Purpose to Meaning.

Spiritus Dei23 Dec 2024 19:37 UTC

−9 points

0 comments6 min readLW link

People aren’t properly calibrated on FrontierMath

cakubilo23 Dec 2024 19:35 UTC

31 points

4 comments3 min readLW link

Near- and medium-term AI Control Safety Cases

Martín Soto23 Dec 2024 17:37 UTC

9 points

0 comments6 min readLW link

[Rationality Malaysia] 2024 year-end meetup!

Doris Liew23 Dec 2024 16:02 UTC

1 point

0 comments1 min readLW link

Printable book of some rationalist creative writing (from Scott A. & Eliezer)

CounterBlunder23 Dec 2024 15:44 UTC

10 points

0 comments1 min readLW link

Monthly Roundup #25: December 2024

Zvi23 Dec 2024 14:20 UTC

18 points

3 comments26 min readLW link

(thezvi.wordpress.com)

Exploring the petertodd / Leilan duality in GPT-2 and GPT-J

mwatkins23 Dec 2024 13:17 UTC

12 points

1 comment17 min readLW link

[Question] What are the strongest arguments for very short timelines?

Kaj_Sotala23 Dec 2024 9:38 UTC

102 points

79 comments1 min readLW link

Reduce AI Self-Allegiance by saying “he” instead of “I”

Knight Lee23 Dec 2024 9:32 UTC

10 points

4 comments2 min readLW link

Funding Case: AI Safety Camp 11

Remmelt, Robert Kralisch and Linda Linsefors

23 Dec 2024 8:51 UTC

60 points

4 comments6 min readLW link

(manifund.org)

What is compute governance?

Vishakha and Algon

23 Dec 2024 6:32 UTC

6 points

0 comments2 min readLW link

(aisafety.info)

Stop Making Sense

JenniferRM23 Dec 2024 5:16 UTC

16 points

0 comments3 min readLW link

Hire (or Become) a Thinking Assistant

Raemon23 Dec 2024 3:58 UTC

139 points

49 comments8 min readLW link

Non-Obvious Benefits of Insurance

jefftk23 Dec 2024 3:40 UTC

21 points

5 comments2 min readLW link

(www.jefftk.com)

Vision of a positive Singularity

RussellThor23 Dec 2024 2:19 UTC

4 points

0 comments4 min readLW link

Ideologies are slow and necessary, for now

Gabriel Alfour23 Dec 2024 1:57 UTC

15 points

1 comment1 min readLW link

(cognition.cafe)

[Question] Has Anthropic checked if Claude fakes alignment for intended values too?

Maloew23 Dec 2024 0:43 UTC

4 points

1 comment1 min readLW link

Vegans need to eat just enough Meat—emperically evaluate the minimum ammount of meat that maximizes utility

Johannes C. Mayer22 Dec 2024 22:08 UTC

55 points

35 comments3 min readLW link

We are in a New Paradigm of AI Progress—OpenAI’s o3 model makes huge gains on the toughest AI benchmarks in the world

garrison22 Dec 2024 21:45 UTC

17 points

3 comments4 min readLW link

(garrisonlovely.substack.com)

My AI timelines

samuelshadrach22 Dec 2024 21:06 UTC

12 points

2 comments5 min readLW link

(samuelshadrach.com)

A breakdown of AI capability levels focused on AI R&D labor acceleration

ryan_greenblatt22 Dec 2024 20:56 UTC

109 points

8 comments6 min readLW link

How I saved 1 human life (in expectation) without overthinking it

Christopher King22 Dec 2024 20:53 UTC

19 points

0 comments4 min readLW link

Checking in on Scott’s composition image bet with imagen 3

Dave Orr22 Dec 2024 19:04 UTC

65 points

0 comments1 min readLW link

Woloch & Wosatan

JackOfAllTrades22 Dec 2024 15:46 UTC

−11 points

0 comments2 min readLW link

A primer on machine learning in cryo-electron microscopy (cryo-EM)

Abhishaike Mahajan22 Dec 2024 15:11 UTC

18 points

0 comments25 min readLW link

(www.owlposting.com)

Notes from Copenhagen Secular Solstice 2024

Søren Elverlin22 Dec 2024 15:08 UTC

9 points

0 comments3 min readLW link

Proof Explained for “Robust Agents Learn Causal World Model”

Dalcy22 Dec 2024 15:06 UTC

28 points

0 comments15 min readLW link

subfunctional overlaps in attentional selection history implies momentum for decision-trajectories

Emrik22 Dec 2024 14:12 UTC

19 points

1 comment2 min readLW link

It looks like there are some good funding opportunities in AI safety right now

Benjamin_Todd22 Dec 2024 12:41 UTC

20 points

0 comments4 min readLW link

(benjamintodd.substack.com)

What o3 Becomes by 2028

Vladimir_Nesov22 Dec 2024 12:37 UTC

149 points

15 comments5 min readLW link

The Alignment Simulator

Yair Halberstadt22 Dec 2024 11:45 UTC

28 points

3 comments2 min readLW link

(yairhalberstadt.github.io)

Theoretical Alignment’s Second Chance

lunatic_at_large22 Dec 2024 5:03 UTC

30 points

3 comments2 min readLW link

Orienting to 3 year AGI timelines

Nikola Jurkovic22 Dec 2024 1:15 UTC

293 points

56 comments8 min readLW link

ARC-AGI is a genuine AGI test but o3 cheated :(

Knight Lee22 Dec 2024 0:58 UTC

3 points

6 comments2 min readLW link

When AI 10x’s AI R&D, What Do We Do?

Logan Riggs21 Dec 2024 23:56 UTC

72 points

17 comments4 min readLW link

AI as systems, not just models

Andy Arditi21 Dec 2024 23:19 UTC

29 points

0 comments7 min readLW link

(andyrdt.com)

Towards a Unified Interpretability of Artificial and Biological Neural Networks

jan_bauer21 Dec 2024 23:10 UTC

2 points

0 comments1 min readLW link