A Note Before the Movie Starts

You are about to read a book about artificial intelligence in medicine. Before you do, I owe you a disclosure.

This book was written in the open — not in the traditional publishing sense, where “open” means the author tweeted about their word count. I mean structurally open. Each chapter was drafted, revised, and published as it was completed, in sequence, on a website that anyone could visit. There were no galley proofs, no embargoed review copies, no six-month lag between manuscript and shelf. The seams are visible. The edits happened in public. If you had been watching from the beginning, you would have seen the book assembling itself, chapter by chapter, like a photograph developing in a darkroom tray.

I should also tell you that AI was involved in the process.

Not in the way you might fear — no language model hallucinated a clinical trial or fabricated a patient story. But AI tools assisted in research, helped refine structure, and contributed to drafting. I am a physician and a former software architect, and I used every tool available to me, the same way I use a stethoscope or a spell-checker or a conversation with a colleague who sees the argument more clearly than I do.

I’m telling you this because the central argument of this book is that AI in medicine must be transparent. That any system influencing clinical decisions should be explainable. That we should know what’s under the hood. It would be a strange kind of hypocrisy to argue for transparency while hiding the conditions of my own work. So here they are: human author, AI tools, public process, real-time publication. The fingerprints of both species are on every page.

This book uses a metaphor that I want to introduce before you encounter it in the chapters themselves, because once you see it, you’ll see it everywhere.

Think about a photograph. A single photograph captures extraordinary detail — the texture of skin, the angle of light, the precise arrangement of objects in a room. It freezes one moment with remarkable fidelity. A physician looking at a single lab result, a single imaging study, a single office visit, is looking at a photograph. And photographs can be beautiful, and diagnostic, and true.

Now play twenty photographs in sequence. Something new appears — something that wasn’t in any individual frame. Motion. A narrative. The way a face changes over weeks. The way a tumor grows or shrinks. The way a patient’s gait deteriorates so slowly that no single visit reveals it, but the sequence makes it unmistakable. This is what AI offers medicine: not better photographs, but the movie. The temporal dimension. The pattern that only emerges when you stop looking at snapshots and start watching the story unfold.

Each chapter of this book is, in its own way, a photograph — a single angle on a vast subject. AI in diagnostics. AI in surgery. AI in drug discovery. AI in mental health. AI in ethics. Individually, they are detailed but incomplete. Read in sequence, I hope they become something more: a moving picture of a profession in transformation, with all the beauty and terror that implies.

A word about who I am, since it matters for what follows.

I am a board-certified vascular neurologist. I have spent years at the bedside — in stroke units, in ICUs, in those 3 AM corridors where the fluorescent lights turn everyone the same shade of gray and the machines speak in monotones. I have held the hand of a patient whose brain was dying in real time. I have told families things that no algorithm will ever be able to say with the necessary gravity.

I am also a former software architect at Microsoft. I have built systems that scaled to millions of users. I have written code that shipped, and code that didn’t, and code that should not have shipped but did anyway because someone made a business decision. I live in both worlds — the clinical and the computational — and I have watched, with increasing urgency, as they collide.

This book is written from that collision point. It is not a manifesto for technologists who want to disrupt healthcare. It is not a lament from a physician who wishes AI would go away. It is an argument — optimistic but not naive, specific but not parochial — that AI will return medicine to its deepest purpose: the care of one human being by another, augmented by machines that handle what machines handle best.

Three principles anchor every chapter:

Augmentation — AI amplifies human judgment; it does not replace it. The physician remains.

Transparency — Any AI influencing a clinical decision must be explainable. Black boxes are unacceptable when lives are at stake.

Equity — AI must reduce health disparities, not encode them. Technology that serves only the privileged is technology that has failed.

If you disagree with any of these, good. Read on. The book will challenge them too.

One more thing. This book is free. It will remain free. I believe that a physician arguing for equitable access to AI-powered medicine should not put that argument behind a paywall. If you find value in what follows, the most useful thing you can do is share it with someone — a colleague, a student, a patient, a policymaker — who might need to hear it.

The lights are dimming. The projector is warming up. Let’s watch the movie.