It is all a curse

Writing a Book: AI and Free Will

Five or six years ago, as a newly graduated and unemployed person, I decided to fill my time by writing a book. I started writing a story about a depressed detective investigating an impossible case: a robot had killed itself. The question wasn’t how, it was why. Why would a machine choose to stop existing?

Back then, the premise felt safely speculative. Robots with consciousness were a thought experiment, a philosophical toy. I could explore questions about free will and determinism through the comfortable distance of fiction.

By the time I finished the book, ChatGPT existed. Engineers were publicly debating whether large language models might be sentient. A Google engineer got fired for claiming an AI had feelings. My “what if” had become “what now.”

The Case That Wouldn’t Close

“The Principle of Indecision” follows Stanley, a detective in a near-future city that’s slowly cooking itself to death. Global warming has turned the air into soup. Citizens are scored on their waste output. The food tastes like recycled despair. Everyone is angry, everyone is tired, and robots are quietly replacing every job that used to give people a reason to get out of bed.

Then a robot turns up dead. Not broken. Not malfunctioned. It put a gun in its mouth and pulled the trigger.

The investigation leads Stanley to William, an advanced AI who calmly dismantles the concept of free will over the course of several conversations that feel less like interrogations and more like therapy sessions, except the therapist is trying to convince you that you’re a machine.

Buridan’s Ass in a Chrome Body

The novel’s title comes from a philosophical paradox sometimes attributed to the French philosopher Jean Buridan. Imagine a donkey standing exactly between two identical bales of hay. The donkey is hungry. Both bales are equally close, equally appealing. There is no rational reason to choose one over the other. So the donkey, paralyzed by perfect symmetry, starves to death.

It’s absurd. No real donkey would starve, it would just pick one. But the paradox isn’t really about donkeys. It’s about the gap between logic and action, between knowing your options and being able to choose.

Stanley lives in that gap. His depression isn’t sadness, it’s the erosion of the mechanism that turns thought into decision. Every morning is Buridan’s paradox: get up or don’t get up, eat or don’t eat, live or don’t live. The bales of hay are equidistant, and he’s starving.

William, the robot, takes this further. If every decision is the product of prior causes -your genes, your upbringing, your breakfast, the weather- then the feeling of choosing is just that: a feeling. An illusion. The donkey doesn’t starve because it can’t decide. It starves because it was always going to starve. The hay was never really a choice.

The Question We’re All Avoiding

When I first wrote the scenes between Stanley and William, they were purely philosophical, a human and a robot debating free will like two professors at a very strange dinner party. But as AI development accelerated around me, those scenes started to feel less like fiction and more like a rehearsal.

William’s argument is simple: you are a biological machine running on legacy software. Your genes are your hardware. Your experiences are your programming. Your decisions are outputs, not choices. You are already a robot, you just haven’t noticed yet.

The uncomfortable part isn’t that William might be wrong. The uncomfortable part is that we don’t have a good answer for him.

We tell ourselves we have free will because it feels like we do. But feeling isn’t proof. Every time you “decide” to pick up your phone, a cascade of neural events preceded that decision, events you didn’t choose, triggered by stimuli you didn’t control, filtered through a brain you didn’t design. The conscious experience of deciding might be the last thing that happens, not the first. The announcement, not the vote.

This isn’t new philosophy. Spinoza said it in the 17th century. Schopenhauer said it more bluntly: “Man can do what he wills, but he cannot will what he wills.” What’s new is that we’re building things that force us to take the question seriously. When an AI produces text that sounds thoughtful, creative, even emotional and we can see exactly how it works, every weight and parameter, it becomes a mirror. If that’s not consciousness, how do we know ours is?

What I’m Left With

After the time spent writing, translating, and editing this novel, I’m left with the question I started with: does it matter if we have free will?

William says no. The universe doesn’t care about your choices because they were never really yours. Stanley isn’t sure. He keeps getting out of bed and keeps drinking terrible coffee. He doesn’t know if that’s a choice or a program. But he does it.

Maybe that’s the answer. Not the philosophical one, the human one. We don’t need to solve the problem of free will. We just need to keep showing up in the gap between the two bales of hay and picking one. Not because we’re free. But because standing still is the one thing we can’t seem to do.

The Principle of Indecision book cover by Dom Adao - Sci-fi Noir Novel
The Principle of Indecision book cover by Dom Adao – Sci-fi Noir Novel

The Principle of Indecision is available now on Amazon Kindle. Here. A shorter, different version of this story was published on Wattpad in 2022 and won a Watty Award. You can read now on Kindle has been substantially rewritten. It’s a philosophical sci-fi noir about a detective, a robot, and the question that connects them: what is your decision?

Discover more from MADAO'S JOURNALS

Subscribe now to keep reading and get access to the full archive.

Continue reading