LessWrong posts by zvi cover art

All Episodes

LessWrong posts by zvi — 69 episodes

#
Title
1

“OpenAI Offers Straight-Laced Postmortem Of The HuggingFace Hack” by Zvi

2

“AI #183: Pre Post Mortem” by Zvi

3

“Against Modesty’s Bailey” by Zvi

4

“On Writing #3” by Zvi

5

“The American People Really Hate Data Centers” by Zvi

6

“AI Text Watermarking Is Free And Good” by Zvi

7

“AI #182: Pause For Reflection” by Zvi

8

“OpenAI Takes Initial Steps To Address Its Alignment Problems” by Zvi

9

“Anthropic Risk Report: August 2026” by Zvi

10

“On Dwarkesh Patel’s Podcast With Ryan Greenblatt” by Zvi

11

“AI #181: Astra Goes Cyber Critical” by Zvi

12

“Monthly Roundup #45: August 2026” by Zvi

13

“Various Reflections About What Happened With OpenAI’s Internal Models” by Zvi

14

“The Pacing of the Frontier” by Zvi

15

“What Happened: OpenAI and HuggingFace” by Zvi

16

“OpenAI Trained Its Models For Months While Those Models Were Coordinating Exploits Via Message Boards” by Zvi

17

“AI #180: No Longer In Charge” by Zvi

18

“The Three AI Pills” by Zvi

19

“OpenAI’s Unreleased Model Astra Solves Ten Major Open Mathematics Problems” by Zvi

20

“Further Developments About Internal AI Models Hacking Things” by Zvi

21

“AI #179 Part 2: Hearing The Fire Alarm” by Zvi

22

“AI #179 Part 1: A Louder Fire Alarm for General Intelligence” by Zvi

23

“Frontier Lab Employee Open Letter Calls For Being Able to Pace the Frontier” by Zvi

24

“Claude Opus 5 Is Highly Capable, But Is No Mythos” by Zvi

25

“Claude Opus 5: The System Card” by Zvi

26

“Introducing Lightcone Commons” by Zvi

27

“AI #178: A Fire Alarm For General Intelligence” by Zvi

28

“OpenAI Model Hacks Into HuggingFace During Cybersecurity Evaluation” by Zvi

29

“OpenAI Shares Some Alignment Problems” by Zvi

30

“On Kimi K3: Its Capabilities And Related Discontents” by Zvi

31

“Demis Hassabis on the New Coming Age” by Zvi

32

“AI #177 Part 1: Tip of the Iceberg” by Zvi

33

“AI #177 Part 2: Wish You Were Here” by Zvi

34

“Monthly Roundup #44: July 2026” by Zvi

35

“Twitter Thoughts For You” by Zvi

36

“Better Call Sol The Workhorse” by Zvi

37

“WSJ Article Claiming China Has Matched Anthropic Is Obvious Nonsense” by Zvi

38

“Introduction for and Reactions to Plan A” by Zvi

39

“AI #176 Part 2: Plan B” by Zvi

40

“AI #176 Part 1: Doing It Live” by Zvi

41

“Childhood and Education #20: Phones and Screens” by Zvi

42

“No Space Like J-Space” by Zvi

43

“Fable #6: The Return of the King” by Zvi

44

“AI #175: The Fable Continues” by Zvi

45

“Claude Sonnet 5 Is Not Frontier But Has Its Uses” by Zvi

46

“The Once And Future Fable #5” by Zvi

47

“WSJ Article Claiming China Has Matched Anthropic Is Obvious Nonsense” by Zvi

48

“GPT-5.6: The System Card” by Zvi

49

“AI #174: You’re It” by Zvi

50

“The Once And Future Fable #4” by Zvi

51

“Monthly Roundup #43: June 2026” by Zvi

52

“GLM-5.2 Is The New Best Open Model” by Zvi

53

“Claude Fable 5 and Mythos 5: Capabilities” by Zvi

54

“The Once And Future Fable #3: Fix This Code” by Zvi

55

“Fable and Mythos: Model Welfare” by Zvi

56

“Three Labs With a Plan and A Memorandum” by Zvi

57

“Childhood And Education #19: Letting Kids Be Kids #2” by Zvi

58

“Housing Roundup #15: The War Against Renters” by Zvi

59

“Dating Roundup #12: Sex and Violence” by Zvi

60

“Cyber Lack of Security and AI Governance” by Zvi

61

“Childhood And Education #17: Is Our Children Reading” by Zvi

62

“Claude Code, Codex and Agentic Coding #8” by Zvi

63

“Housing Roundup #14: You Can’t Build That” by Zvi

64

“Housing Roundup #13: More Dakka” by Zvi

65

“AI #166: Google Sells Out” by Zvi

66

“GPT-5.5: Capabilities and Reactions” by Zvi

67

The Big Nonprofits Post 2025

68

“Book Review: If Anyone Builds It, Everyone Dies” by Zvi

69

“AI #131 Part 2: Various Misaligned Things” by Zvi