AI Poli­tics: Po­lariza­tion and Chaos

American PsychohistoryMar 31, 2025, 11:46 PM
4 points
0 comments4 min readLW link

Call for Col­lab­o­ra­tion: Renor­mal­iza­tion for AI safety

Lauren GreenspanMar 31, 2025, 9:01 PM
35 points
0 comments4 min readLW link

A re­sponse to OpenAI’s “How we think about safety and al­ign­ment”

HarlanMar 31, 2025, 8:58 PM
11 points
0 comments6 min readLW link
(intelligence.org)

Op­por­tu­nity Space: Renor­mal­iza­tion for AI Safety

Lauren GreenspanMar 31, 2025, 8:55 PM
22 points
0 comments6 min readLW link

Renor­mal­iza­tion Roadmap

Lauren GreenspanMar 31, 2025, 8:34 PM
62 points
7 comments18 min readLW link

AISN #50: AI Ac­tion Plan Responses

Mar 31, 2025, 8:13 PM
4 points
0 comments6 min readLW link
(newsletter.safe.ai)

On Down­votes, Cul­tural Fit, and Why I Won’t Be Post­ing Again

funnyfrancoMar 31, 2025, 7:26 PM
0 points
32 comments2 min readLW link

Fundrais­ing for Mox: cowork­ing & events in SF

Austin ChenMar 31, 2025, 6:25 PM
27 points
0 commentsLW link
(manifund.org)

OpenAI #12: Bat­tle of the Board Redux

ZviMar 31, 2025, 3:50 PM
141 points
1 comment9 min readLW link
(thezvi.wordpress.com)

Rou­tine Novelty

BazingaBoyMar 31, 2025, 3:47 PM
1 point
0 comments1 min readLW link

Why does Claude Speak Byzan­tine Mu­sic No­ta­tion?

Lennart FinkeMar 31, 2025, 3:13 PM
18 points
2 comments3 min readLW link

When the Wannabe Rambo Co­me­dian Cried

P. JoãoMar 31, 2025, 2:47 PM
32 points
0 comments3 min readLW link

A Frac­tion of Global Mar­ket Cap­i­tal­iza­tion as the Best Currency

Greenless MirrorMar 31, 2025, 1:30 PM
1 point
25 comments7 min readLW link

The Apoca­lypse is Near. Can Hu­man­ity Coex­ist with Ar­tifi­cial Su­per­in­tel­li­gence?

Jakub GrowiecMar 31, 2025, 1:17 PM
4 points
0 comments11 min readLW link

Sam Alt­man’s sister claims Sam sex­u­ally abused her—Part 11: List of An­nie’s on­line ac­counts, References

pythagoras5015Mar 31, 2025, 12:26 PM
3 points
1 comment105 min readLW link

Sam Alt­man’s sister claims Sam sex­u­ally abused her—Part 10: re­sponses from Sam and his fam­ily mem­bers; my perspective

pythagoras5015Mar 31, 2025, 12:26 PM
1 point
1 comment25 min readLW link

Sam Alt­man’s sister claims Sam sex­u­ally abused her—Part 9: liter­a­ture on child sex­ual abuse and trauma

pythagoras5015Mar 31, 2025, 12:25 PM
3 points
0 comments141 min readLW link

Sam Alt­man’s sister claims Sam sex­u­ally abused her—Part 6: Timeline, continued

pythagoras5015Mar 31, 2025, 12:25 PM
3 points
0 comments88 min readLW link

Sam Alt­man’s sister claims Sam sex­u­ally abused her—Part 3: Timeline, continued

pythagoras5015Mar 31, 2025, 12:24 PM
3 points
0 comments81 min readLW link

Sam Alt­man’s sister claims Sam sex­u­ally abused her—Part 2: An­nie’s law­suit; the re­sponse from Sam, his broth­ers, and his mother; Timeline

pythagoras5015Mar 31, 2025, 12:24 PM
3 points
0 comments65 min readLW link

Story Feed­back Re­quest: The Policy—Emer­gent Align­ment, Re­cur­sive Cog­ni­tion, and AGI Trajectories

queeliusMar 31, 2025, 11:08 AM
10 points
2 comments48 min readLW link

On the Im­pli­ca­tions of Re­cent Re­sults on La­tent Rea­son­ing in LLMs

Rauno ArikeMar 31, 2025, 11:06 AM
34 points
6 comments13 min readLW link

OpenAI lost $5 billion in 2024 (and its losses are in­creas­ing)

RemmeltMar 31, 2025, 4:17 AM
26 points
15 comments12 min readLW link
(www.wheresyoured.at)

The Leapfrog­ging Ter­minus and the Fuzzy Cut

Jim PivarskiMar 31, 2025, 4:08 AM
22 points
6 comments13 min readLW link

CoreWeave Is A Time Bomb

RemmeltMar 31, 2025, 3:52 AM
5 points
0 comments2 min readLW link
(www.wheresyoured.at)

Down­stream ap­pli­ca­tions as val­i­da­tion of in­ter­pretabil­ity progress

Sam MarksMar 31, 2025, 1:35 AM
112 points
3 comments7 min readLW link

Effi­ciency as a 2-place word

Adam ZernerMar 31, 2025, 1:17 AM
12 points
2 comments6 min readLW link

Mee­tups Notes (Q1 2025)

jennMar 31, 2025, 1:12 AM
30 points
2 comments8 min readLW link

Ap­par­ent In­tro­spec­tion in Claude: A Case Study in Pro­jected Mind

robert_saltzmanMar 31, 2025, 12:51 AM
5 points
0 comments1 min readLW link

Align­ment First, In­tel­li­gence Later

Chris LakinMar 30, 2025, 10:26 PM
18 points
5 comments3 min readLW link

[Question] Why do many peo­ple who care about AI Safety not clearly en­dorse PauseAI?

humnrdbleMar 30, 2025, 6:06 PM
45 points
42 comments2 min readLW link

Enu­mer­at­ing ob­jects a model “knows” us­ing en­tity-de­tec­tion fea­tures.

Alex GibsonMar 30, 2025, 4:58 PM
6 points
2 comments6 min readLW link

Bonn ACX Meetup Spring 2025

Fernand0Mar 30, 2025, 3:12 PM
2 points
1 comment1 min readLW link

What does al­ign­ing AI to an ide­ol­ogy mean for true al­ign­ment?

StanislavKrymMar 30, 2025, 3:12 PM
1 point
0 comments8 min readLW link

How to en­joy fail at­tempts with­out self-de­cep­tion (tech­nique)

YanLyutnevMar 30, 2025, 1:49 PM
9 points
0 comments9 min readLW link

Me­mory Per­sis­tence within Con­ver­sa­tion Threads with Mul­ti­modal LLMS

sjay8Mar 30, 2025, 7:16 AM
4 points
0 comments1 min readLW link

How I talk to those above me

Maxwell PetersonMar 30, 2025, 6:54 AM
102 points
16 comments8 min readLW link

How do SAE Cir­cuits Fail? A Case Study Us­ing a Starts-with-‘E’ Let­ter De­tec­tion Task

adsingh-64Mar 30, 2025, 12:47 AM
1 point
0 comments3 min readLW link

Climb­ing the Hill of Experiments

nomagicpillMar 29, 2025, 8:37 PM
4 points
0 comments6 min readLW link
(nomagicpill.github.io)

[Question] Does the AI con­trol agenda broadly rely on no FOOM be­ing pos­si­ble?

Noosphere89Mar 29, 2025, 7:38 PM
22 points
3 comments1 min readLW link

Ex­er­cis­ing Rationality

EggsMar 29, 2025, 7:08 PM
4 points
0 comments4 min readLW link

Yeshua’s Basilisk

Alex BeymanMar 29, 2025, 6:11 PM
8 points
1 comment4 min readLW link

AI Needs Us? In­for­ma­tion The­ory and Hu­mans as data

tomdekanMar 29, 2025, 3:51 PM
0 points
6 comments4 min readLW link

Auto Shut­down Script

jefftkMar 29, 2025, 1:10 PM
16 points
5 comments1 min readLW link
(www.jefftk.com)

Pro­posal for a Post-La­bor So­cietal Struc­ture to Miti­gate ASI Risks: The ‘Game Cul­ture Civ­i­liza­tion’ (GCC) Model

Beyond SingularityMar 29, 2025, 11:31 AM
2 points
0 comments4 min readLW link

Tor­ment­ing Gem­ini 2.5 with the [[[]]][][[]] Puzzle

CzynskiMar 29, 2025, 2:51 AM
48 points
36 comments3 min readLW link

Sin­gu­lar­ity Sur­vival Guide: A Bayesian Guide for Nav­i­gat­ing the Pre-Sin­gu­lar­ity Period

mbrooksMar 28, 2025, 11:21 PM
6 points
4 comments2 min readLW link

Soft­max, Em­mett Shear’s new AI startup fo­cused on “Or­ganic Align­ment”

Chris LakinMar 28, 2025, 9:23 PM
59 points
1 comment1 min readLW link
(www.corememory.com)

The Pando Prob­lem: Re­think­ing AI Individuality

Jan_KulveitMar 28, 2025, 9:03 PM
128 points
14 comments13 min readLW link

Selec­tion Pres­sures on LM Personas

Raymond DouglasMar 28, 2025, 8:33 PM
30 points
0 comments3 min readLW link