U = I x R^2: how the book presents the equation

4 min read · 771 words
Share:
Michael Darius Eastwood
Michael Darius Eastwood · Independent AI alignment researcher
Published
Michael Darius Eastwood · Book companion · 3 July 2026
Michael Darius Eastwood, independent researcher, London: originator of the embedded-correction alignment thesis (manuscript 8 December 2024, SHA-256 anchored: f0d1f38f).

The equation sits at the centre of the book's introduction, presented with more care than a slogan.

At the centre of this book is something I call the ARC Principle, expressed in a formula that might seem deceptively simple: U = I x R^2. Universe equals Intelligence multiplied by Recursion squared.Infinite Architects, Introduction

The R-squared distinction

The book is explicit that the exponent is not decoration. The distinction between U = I x R and U = I x R^2 is described as crucial. Linear recursion adds; squared recursion multiplies. Each iteration does not just add to what came before; it multiplies the capacity for future iterations. The book's image is walking up a hill versus riding a rocket: evolution took billions of years to produce human intelligence, human culture compressed that into millennia, artificial intelligence at computational speeds might compress millennia into decades, perhaps less.

What the equation is, and what it is not

The book is careful to mark its own status. The ARC Principle is described as a framework, not yet a proven physical law, and the author asks the reader to think of it as Darwin's "descent with modification" was before population genetics: a lens that organises observations and generates predictions, even before we have the tools to quantify it precisely. Whether it will be formalised into something calculable or will remain a conceptual tool like "survival of the fittest", the author says, he cannot say. What he can say is that it keeps proving useful.

Reading the equation across the evidence chain

The equation has a slightly longer history than the book alone. The December 2024 manuscript stated U = I x R. The subsequent papers formalise this as U = I x R^alpha, where alpha is a recursion-form-dependent exponent. The published book presents U = I x R^2, that is, the alpha equals 2 case, as the featured formulation. The book's job is not to prove alpha; that work belongs to the papers. Its job is to make the presented case, alpha equals 2, legible to a general reader and to show what follows if it is even approximately right.

Why the book compares to E = mc^2

The chapter draws a careful parallel with Einstein's equation. Einstein's formula transformed the understanding of what matter and energy fundamentally are. The book proposes something similarly transformative about intelligence and creation: not that the equation is proven in the way Einstein's has been proven, but that the parallel is instructive. It suggests what kind of truth the ARC Principle might be claiming, not just a useful analogy but a statement about how reality is structured, if the framework survives its own tests. The book is explicit that this is a comparison of ambition and shape, not a claim to equivalent standing.

What the R-squared connects to

Later chapters return to the exponent's geometry. The book's argument is that squared terms in physics are not decorations; they are load-bearing features of how reality works. If the R^2 in the ARC Principle is of the same character, then recursion is woven into the geometry of existence. The book's framing is careful: this is a suggestion about structural affinity, not a proof.

The equation's history is deliberately preserved in the programme's canonical facts register: U = I x R appears in the December 2024 manuscript (V2, lines 6-9); Papers I and II formalise it as U = I x R^alpha; the published book presents the alpha equals 2 case. The three forms are not competing claims. They are one family cited alongside its artifacts.

From the book Infinite Architects: Intelligence, Recursion, and the Creation of Everything by Michael Darius Eastwood.

Buy on Amazon UK Amazon US

Stay informed

New posts on AI alignment, convergence evidence, and the ARC/Eden research programme.

Get updates →