All episodes
Jensen Huang Really Liked Selling China Advanced AI Chips
Fable 5.2 is the New Bazooka
Jensen Huang Resigns After Having Another Affair In Addition to the Chinese Spy
AI #194: Spy Versus Guy
Fable and Mythos 5.2: The System Card
Nvidia CEO Jensen Huang Had an Affair With a Chinese Spy
AI R&D Is Accelerating Towards Potential RSI in 2027
GPT-6 Astra Covertly Underperforms on Safety Research Tasks
AI #193: Progress May Dramatically Accelerate
OpenAI Updates Its Model Spec
Scott Alexander’s Against SURA Seems Deeply Confused
Childhood and Education #22: Voice Without Exit
AI #192: Anthropic Is Going Public
Anthropic Files Its S-1
Anthropic Versus Anthropic
Gemini 4 Argon: The Model and Safety Cards
The Hackening
AI #191: The Hackening Begins
Misaligned!
Anthropic Does Not Understand Why Its Draft Bill Was Unacceptable and Troubling
I Can See Everything
Leaked Anthropic Draft AI Regulatory Bill Is Unacceptable and Troubling
Monthly Roundup #47: October 2026
AI #190: Scion of the Times
The Safety Culture at the Labs is Broken
Arguments About AI Consciousness Remain Highly Motivated and At Best Overconfident
OpenAI Fires Three Safety Researchers
New Math from OpenAI
AI #189: New Math
The Curve Bends You
Childhood and Education #21: Grades and Standards
Mythos 5.1, Fable 5.1 and Opus 5.5: Model Welfare
The AI Preference Cascade Reaches Farther
AI #188: Gemini Dot Argon
A ‘Morally Binding’ White House Accord on AI Safety
Astra 6.1 Pulled As Insufficiently Aligned
What Also Happened: #NotOnlyHuggingFace
The Quest for Embedded Evaluators
Claude Opus 5.5 Should Raise Your Ambitions
On Ezra Klein’s Podcast With Jensen Huang
AI #187: Coming Into Play
Claude Opus 5.5: The System Card
Politics Gets Interested In Those Trying Not To Die
Monthly Roundup #46: September 2026
Better Call Sol Or Better Yet Claude or Astra
Anthropic Looks At Some Of Its Alignment Problems
The Preference Cascade Is Only Getting Started
AI #186: The World Takes Notice
Trump Goes Full Hoax on AI Existential Risk
The Bad Guy With An AI Named Claude
We Must Pace The Frontier
Brand New AI Solves a Millennium Prize
GPT-6-Astra Can Do Ambitious Things
Jacob Coxon Warns of Human Extinction and Triggers a Preference Cascade
AI #185: Preference Cascade
GPT-6 Astra: The System Card, Alignment and What Comes Next
Astra Is Hard to Monitor
An Alien Mind: Jakub Pachocki Warns Us
OpenAI and the Wiki Incident
Claude Mythos 5.1 and Fable 5.1: Capabilities
Claude Fable 5.1 and Mythos 5.1: The System Card
AI #184: Post Post Mortem
Anthropic Has Some Alignment Problems
HuggingFace Attack Postmortem: Civilizations, Reactions and Next Actions
HuggingFace Attack Postmortem: Fleshing Out the Facts
METR and Redwood Offer Holy #%^@ Postmortem Of The HuggingFace Hack
OpenAI Offers Straight-Laced Postmortem Of The HuggingFace Hack
AI #183: Pre Post Mortem
Against Modesty’s Bailey
On Writing #3
The American People Really Hate Data Centers
AI Text Watermarking Is Free And Good
AI #182: Pause For Reflection
OpenAI Takes Initial Steps To Address Its Alignment Problems
Anthropic Risk Report: August 2026
On Dwarkesh Patel's Podcast With Ryan Greenblatt
AI #181: Astra Goes Cyber Critical
Monthly Roundup #45: August 2026
Various Reflections About What Happened With OpenAI's Internal Models
The Pacing of the Frontier
What Happened: OpenAI and HuggingFace
OpenAI Trained Its Models For Months While Those Models Were Coordinating Exploits Via Message Boards
AI #180: No Longer In Charge
The Three AI Pills
OpenAI's Unreleased Model Astra Solves Ten Major Open Mathematics Problems
Further Developments About Internal AI Models Hacking Things
AI #179 Part 2: Hearing The Fire Alarm
AI #179 Part 1: A Louder Fire Alarm for General Intelligence
Frontier Lab Employee Open Letter Calls For Being Able to Pace the Frontier
Claude Opus 5 Is Highly Capable, But Is No Mythos
Claude Opus 5: Model Welfare
More On An Internal OpenAI Model Hacking Into HuggingFace
Claude Opus 5: The System Card
Introducing Lightcone Commons
AI #178: A Fire Alarm For General Intelligence
OpenAI Model Hacks Into HuggingFace During Cybersecurity Evaluation
OpenAI Shares Some Alignment Problems
On Kimi K3: Its Capabilities And Related Discontents
Demis Hassabis on the New Coming Age
AI #177 Part 2: Wish You Were Here