Stephen Wolfram is a British-American scientist and entrepreneur who has spent five decades on a single bet: that computation — not mathematics, and not experiment alone — is the deepest framework for understanding the world. He is the founder and CEO of Wolfram Research, the creator of Mathematica, Wolfram|Alpha, and the Wolfram Language, the author of A New Kind of Science, and the originator of the Wolfram Physics Project. Nearly everything he has built, including the company, exists to fund and equip that bet.
Formation: the prodigy who left early
Born in London on August 29, 1959 — his father a textile businessman and novelist, his mother an Oxford philosophy professor — Wolfram was publishing particle-physics papers at fifteen. His schooling was elite and mostly ornamental: a King's Scholar at Eton, then St John's College, Oxford, where he found the lectures not worth attending, worked out of the Nuclear Physics building, and left in 1978 without a degree. By then he had published roughly ten papers. At Caltech he took a PhD in theoretical physics in 1979 at age twenty — his committee chaired by Richard D. Field and including Richard Feynman — and joined the faculty.
At Caltech he led the construction of SMP, the first modern computer algebra system, and fought the administration over who owned it. In June 1981, at twenty-one, he became the youngest recipient of a MacArthur Fellowship. That same year he made the turn that defines him: away from particle physics and toward the origins of complexity, using the humblest objects he could find — cellular automata — as material for computer experiments.
Building the instruments
His 1983 Reviews of Modern Physics paper, "Statistical mechanics of cellular automata," founded much of the field: it sorted elementary automata into four behavioral classes and treated them as objects of empirical study. Through the mid-1980s — at the Institute for Advanced Study, then the University of Illinois — he built the scaffolding of what he called complex systems research: the field's first journal, Complex Systems, and its first research center. Then he left academia.
Wolfram Research was founded in 1987; Mathematica 1.0 shipped on June 23, 1988. He has been unusually candid about the motive: he wanted the instrument for himself — "a bit like Galileo got to use his telescope," as he put it at TED — pointed not at the astronomical universe but at the computational one. Mathematica, then Wolfram|Alpha (launched May 2009), then the Wolfram Language (2014) form one continuous stack: an effort to make knowledge itself computable, expressed in symbolic form rather than searched.
A New Kind of Science
From roughly 1991 to 2002, working mostly at night and largely in seclusion, he wrote A New Kind of Science — nearly 1,200 pages, published by his own imprint on May 14, 2002, an immediate bestseller and an immediate controversy. Its core claim is empirical: simple programs, iterated, routinely produce behavior of arbitrary complexity. Rule 30 — a few lines long — generates randomness good enough to seed a practical generator still in use. From there the book builds outward: the four classes; computational irreducibility, the claim that for many systems the only way to know what they do is to run them; and the Principle of Computational Equivalence, the claim that nearly everything above a low complexity threshold computes at the same maximal level — weather, brains, and Turing machines alike.
The deeper break is methodological: computer experiments as first-class science, set deliberately against the theorem-and-proof tradition. Critics called the book grandiose and light on credit to prior work; supporters called it a founding document. Both readings remain in the record.
The physics project and the ruliad
On April 14, 2020, he announced the Wolfram Physics Project: the hypothesis that the universe is a hypergraph rewriting system — space the graph, time the succession of rewrites — with general relativity and quantum mechanics emerging for observers like us from causal structure and multiway branching. The project runs as unusually open science: papers, tools, working materials dating to the 1990s, a Registry of Notable Universes, and hundreds of hours of livestreamed research.
From it came the ruliad, his name for the entangled limit of all possible computations — every rule on every initial condition, run forever. On his account there is no "why this rule" left to answer: the universe runs all rules, and bounded observers like us sample a slice whose robust features look like physics. Whether this is a candidate theory of everything or a reframing of known structure is the live question; no rule for our universe has been identified, and the correspondences published so far are formal rather than empirical.
The instrumented self
The same engineering reflex applies to the person. Since the late 1980s he has archived nearly everything measurable about his own life — a third of a million sent emails, more than a hundred million keystrokes, meetings, calls, steps — and published the analysis as "personal analytics." His productivity is similarly instrumented: decades of "personal infrastructure" design, remote management of a company of several hundred people largely by screen, days structured around deep work. In February 2023 his essay "What Is ChatGPT Doing … and Why Does It Work?" became the most-read public explanation of large language models and a short book; within weeks Wolfram|Alpha was a ChatGPT plugin — the symbolic engine lending the statistical one what he calls computational superpowers.
What the record does not settle
The record is rich in first-person material — Wolfram is an unusually thorough self-documenter — and thin exactly where it matters: the long-run standing of A New Kind of Science, the empirical status of the physics project, even the company's founding year (1986 in the MacArthur biography, 1987 in his own telling). The archive figures are self-reported; the strongest claims are his own. This index preserves those seams rather than smoothing them over.
This index was compiled from public sources and does not imply the subject's endorsement. Citations live in the packet's source catalog.