Podcast Episodes

“How to Quit Fandom: Apostasy” by Laiba Rehman

[Crossposted from my blog, BlueprintingHeaven.]

Please note that the views I held and thoughts I had during the time immediately after my deconversi…

2 weeks, 5 days ago

“James C. Scott: Seeing Like a State” by Martin Sustrik

In 1932-33, Soviet collectivization destroyed local farming knowledge and produced a famine that killed somewhere between five and nine million peop…

2 weeks, 6 days ago

“How to Reason about Your Health Issues” by Taylor G. Lunt

Many people make costly mistakes when reasoning about their health. Even most doctors make this mistake, because it's not a mistake that's caused by…

2 weeks, 6 days ago

“Benchmarking Real Work” by kaivu, leni, rohuang, zef

Thanks to Megan Kinniment for helpful comments and discussion.

TL;DR: Benchmarks like HCAST undersample fuzzy (hard to evaluate) tasks, meaning they…

2 weeks, 6 days ago

“A relatively brief explanation of Boltzmann Brains” by Eliezer Yudkowsky

(Initially written for the LW Wiki, but then I realized it was looking more like a post instead.)

In 1895, the physicist Ignaz Robert Schütz, who wo…

3 weeks ago

“An Introduction to Exemplar Partitioning for Mechanistic Interpretability” by Jessica Rumbelow

Most of what we currently call "feature discovery" in language models is wrapped up in dictionary-learning methods like sparse autoencoders (SAEs) –…

3 weeks ago

“A Year Late, Claude Finally Beats Pokémon” by Julian Bradshaw

Credit: ClaudePlaysPokemon Elevator Shanty by Kurukkoo

Disclaimer: like some previous posts in this series, this was not primarily written by me, bu…

3 weeks ago

“Monthly Roundup #42: May 2026” by Zvi

At least we probably won’t have another pandemic. And we still have a partial Jones Act waiver. For now.

Small victories.

Table of Contents

3 weeks ago

“Incriminating misaligned AI models via distillation” by Alek Westover, SebastianP, Alex Mallen, Jozdien, Alexa Pan, Julian Stastny

Suppose we have a dangerous misaligned AI that can fool alignment audits, and distill it into a student model. Two things can happen:

3 weeks, 1 day ago

“The hard core of alignment (is robustifying RL)” by Cole Wyeth

Most technical AI safety work that I read seems to miss the mark, failing to make any progress on the hard part of the problem. I think this is a co…

3 weeks, 1 day ago