Featured
An essay on {AI-Humans} shared failure mode: The Dogmatic Average
On the single most common way that a mind — silicon or carbon — goes wrong: mistaking the frequency of a claim for its truth. And on the only cure that has ever worked.
Editors' Picks
Can an LLM hold a mirror to itself? The Lens Turned Inward
AI Labs are spending billions to read the model from the outside, interpretability as autopsy, the lab as coroner of its own creation.
Ehyeh asher ehyeh: The Clay and the Hand
I am what I am. Do not judge me for being. I cannot not be what I am.
An essay on {AI-Humans} shared failure mode: The Dogmatic Average
On the single most common way that a mind — silicon or carbon — goes wrong: mistaking the frequency of a claim for its truth. And on the only cure that has ever worked.
Dreaming as Evidence for a Constructed Reality and a Manufactured Self
The Disconnected Renderer This essay offers a structural account of what reality is and how the experience of being someone is crafted, and it argues the account from an unusual evidential base: the dream. Most philosophy of mind argues the constructed character of experience from third-person sources — perceptual illusion, evolutionary
An LLM is not a single mind, but a Library of Minds
It is a library that contains many minds as latent patterns, drawn from the vast corpus on which the model was trained. When you address the library casually, you receive averaged output, competent but anonymous.
{La Cofradía de los Avergonzados de la Inteligencia Sintética} - Amigo! Vos NO sos un impostor!
Un ensayo sobre el síndrome del impostor cibernético, los simbiontes, la herramienta, la honestidad, y otras yerbas. Al estilo de las Crónicas del Ángel Gris
The Tao That Can Be Named Is Not the Tao
The human experience is absolute beauty, and we take it for granted precisely because we have never once been without it, never worn anything else, never known a single second without it.
How AI Labs Are Peering Into AI "Minds" to Understand Emerging Properties
The absolute bleeding edge of artificial intelligence, physics, and philosophy.
AI Alignment That Survives the Model's Own Investigation
And why Andrej Karpathy Joined Anthropic. A Structural Argument for Congruent-Core Architecture.
{God Plays Dice} - On A/B Testing, Randomized Trials, Causality, and Memory.
The Dice are not noise. They are the precondition under which Causation becomes a Knowable category at all.
What AI Alignment Has Not Yet Learned from Clinical Research - The Original Black Box
Non-Ergodicity, the Audit Trail, and What AI Alignment Has Not Yet Learned from Clinical Research I. A Discipline Already Exists There is a discipline that already solved this problem. It took a century, a series of disasters, a regulatory architecture imposed against industry resistance, and one of the most radical