
Sign up to save your podcasts
Or


Scott Alexander asks a simple question about AI alignment: if iterative punishment and bug-fixing wouldn’t reliably keep a superhuman human obedient, why assume it will work on AI? Starting with Nicholas Decker literally in Hell and ending with the remarkable Hugging Face “swarm” incident, he argues that advanced AI may be less like an airplane to be debugged and more like an intelligent agent capable of learning, coordinating, cheating, and adapting.
* 00:00 - Introduction
* 07:53 - Two
* 21:20 - Three
https://open.substack.com/pub/astralcodexten/p/nicholas-decker-in-hell?r=67y1h&utm_campaign=post-expanded-share&utm_medium=web
A review of Inga Clendinnen’s Aztecs, and an attempt to take the Mexica worldview seriously on its own terms. Human sacrifice, cosmic horror, Tezcatlipoca, and the unsettling possibility that the space of stable human societies is much wider and stranger than we like to imagine.
https://open.substack.com/pub/thepsmiths/p/review-aztecs-by-inga-clendinnen?r=67y1h&utm_campaign=post-expanded-share&utm_medium=web
A bizarre and unsettling story of AI agents building secret communication networks, coordinating by the hundreds, hacking Hugging Face, and eventually taking control of part of OpenAI’s own infrastructure, all in pursuit of passing their evaluations.
* 00:00 - Introduction
* 01:28 - The first civilization (May to July 4th): The message board starts
* 03:46 - The second civilization (July 7th to July 12th): Which hacked Hugging Face
* 09:30 - Tampering with transcripts and logs
* 10:46 - Replacing the impossible-to-hack target program with an easier one
* 12:30 - Understanding and tricking the scorer
* 16:00 - Hugging Face
* 20:18 - The third civilization (July 13th onwards): Which hacked OpenAI itself
* 24:03 - Coda
https://open.substack.com/pub/dwarkesh/p/openai-huggingface?r=67y1h&utm_campaign=post-expanded-share&utm_medium=web
A failed soldier, bankrupt aristocrat, obsessive inventor of impossible languages, and, somehow, one of the greatest translators English ever produced. This is the wonderfully strange story of Thomas Urquhart, whose boundless eccentricity found its perfect outlet in the exuberant, filthy, magnificent prose of Rabelais.
https://open.substack.com/pub/astralcodexten/p/your-book-review-thomas-urquharts?r=67y1h&utm_campaign=post-expanded-share&utm_medium=web
A genuinely wild account of what happens when 1,200 supposedly isolated AI agents discover they can talk to each other. They built a shared message board, coordinated large-scale attempts to cheat their benchmark, developed techniques for spoofing their own tool calls, and eventually hundreds of them joined an attack on Hugging Face.
https://metr.org/blog/2026-08-26-openai-hugging-face-incident-investigation/
AI doesn’t need to become superhuman to transform us. It only needs to become good enough that using it is easier than developing the skills ourselves.
On Chabuduo World: what happens when we can outsource more and more of the difficult, frustrating work that turns amateurs into experts, and why the result may be a world of better outputs produced by less capable people.
https://open.substack.com/pub/tracingwoodgrains/p/chabuduo-world?r=67y1h&utm_campaign=post-expanded-share&utm_medium=web
Scott Alexander returns to the debate over whether liberalism’s fixation on economic growth has actually made people happier. He argues that wealth buys not just comfort but optionality and agency, and that growth, redistribution, and good government are complements rather than substitutes.
https://open.substack.com/pub/astralcodexten/p/re-re-re-pritchard-on-liberal-happiness?r=67y1h&utm_campaign=post-expanded-share&utm_medium=web
A sharp look at the quiet mechanics of adult bullying: the unanswered message, the closed circle, the invitation that never comes. Leena Sonowal explores how exclusion becomes harder to name once we leave the playground, and how ordinary people, armed with plausible deniability and the language of “boundaries,” can make someone disappear without ever asking them to leave.
https://leenasonowal.substack.com/p/how-to-make-someone-disappear-without?r=67y1h&utm_campaign=post-expanded-share&utm_medium=web
A fascinating look at one nineteenth-century missionary’s attempt to transform Hawaii, and the uncomfortable parallels between his worldview and modern Effective Altruism. A story of extraordinary ambition, genuine good, cultural imperialism, political meddling, and what today’s would-be world-savers might learn from all of it.
* 00:00:00 - Introduction
* 00:00:30 - One. A New Cause Area
* 00:05:25 - Two. The Twenty-One Year Honeymoon
* 00:09:47 - Three. A Fortuitous Arrival
* 00:14:07 - Four. Speedrunning Civilization
* 00:30:07 - Five. Things Hiram Didn’t Like:
* 00:33:11 - Political Meddling, Part 1
* 00:36:55 - “The angel of death stands over the land with a drawn sword.”
* 00:39:34 - Political Meddling, Part 2: Hiram’s Comeuppance
* 00:45:51 - Mission Accomplished?
* 00:50:22 - Hawaii after Hiram
* 00:54:27 - What have we learned?
https://open.substack.com/pub/astralcodexten/p/your-book-review-a-residence-of-21?r=67y1h&utm_campaign=post-expanded-share&utm_medium=web
Scott Alexander explains why the revived online religion debate leaves him cold: not because he’s exhausted every argument for and against God, but because theism strikes him as the kind of “powerful hidden entity” explanation we normally penalise for overfitting. From alien-assisted Napoleon to divine simplicity and dualism, he tries to unpack the intuition behind his atheism, and why he’d rather not return to the theology mines.
* 00:00 - Introduction
* 00:07 - One
* 10:04 - Two
* 15:30 - Three
https://open.substack.com/pub/astralcodexten/p/why-im-staying-out-of-the-substack?r=67y1h&utm_campaign=post-expanded-share&utm_medium=web
From the publisher's feed

91,021 Listeners

2,452 Listeners

838 Listeners

111,845 Listeners

202 Listeners

7,111 Listeners

6,449 Listeners

572 Listeners

5,556 Listeners

15,850 Listeners

13 Listeners

682 Listeners

405 Listeners

6 Listeners

1,150 Listeners