Philosophy
Philosophy has stayed with me as an interest alongside linguistics. This page is for writing that is not research: essays that think a question through rather than measure it. They make no claim to academic standing in philosophy.
The AI Is Only Human
Why a Superintelligence May Not Kill Us · an essay, November 2026 · 29 pages
Suppose a machine one day understands the world, and us, far better than we do. Would that give it the right to decide what happens to us? The essay starts from a comparison Yuval Noah Harari made in an interview: that we might come to stand to a superior intelligence as cattle stand to us. Its answer is a narrow one. Being better at working things out settles what follows from one's ends, not which ends to have, and so greater capability alone confers no authority over others. It offers no proof that a superintelligence will spare us, and says so.
The essay is built like a tower of floors, from a short parable about a machine and an old man in Athens up to a final letter addressed to whoever, or whatever, is reading. A thin column in the margin tells the whole argument as it might be told to a child of seven, as a check on the formal version. Short fragments of a machine's drafts run between the chapters.
How it was written. The essay was written in continuous dialogue with language models, and its subject is partly its own making. The question, the central ideas, the thought-fragments and metaphors, the selection and rejection of arguments, and all editorial decisions are mine. The machines drafted most of the English sentences, proposed structure and counter-arguments, helped check the references and generated the illustrations to my specification. The title page deliberately carries no author's name; the colophon explains why, and the final pages set out the division of labour in full. An adversarial examination by a language model is not peer review. Comments and objections are most welcome: research@woidwide.net.