April 2026
The Laziness of Thinking: An Introduction to My Upcoming Book
An introduction to the ideas behind my forthcoming essay — and the question of what we lose when intelligent machines think for us.
Read →Essays & Reflections
On artificial intelligence, cognition, philosophy of technology, and the questions that don't fit in a research paper.
An introduction to the ideas behind my forthcoming essay — and the question of what we lose when intelligent machines think for us.
Read →The mechanics of next-token prediction turn out to be a remarkably accurate model of how humans use cognitive shortcuts.
Read →Silicon Valley's deepest assumption is that removing friction is always good. It may be making us less capable of careful thought.
Read →We started evaluating AI on bar exams and medical certifications. Now our models ace those tests. The measurement problem runs deeper than it looks.
Read →The Epoch AI study has a quiet implication: we are approaching the limits of human-generated training data. What comes next?
Read →The difference between a tool that waits for a prompt and one that works through the night is not just technical.
Read →The difference between consuming an AI via a closed API and owning its weights is not just technical — it is a question of cognitive sovereignty.
Read →Two schools of thought, two visions of intelligence. The scaling laws camp and the world models camp disagree on something fundamental.
Read →Jacques Ellul wrote about autonomous technique in 1954. Reading him now, in the age of self-optimising AI systems, feels less like history and more like prophecy.
Read →The Last Man wants comfort and safety above all else. We may be building the infrastructure for his perfect world.
Read →The journey from "king − man + woman = queen" to the dense, contextual representations inside a modern transformer is a story about what it means to represent meaning in a machine.
Read →