The Self Alignment Framework (SAF) did not begin as an AI project, or with any intention beyond my own curiosity and quest for truth.

I grew up in a small village in El Salvador, with little access to formal education, but with an endless curiosity about nature and how things fit together. At sixteen, when I moved to the United States, I discovered philosophy. Plato, Aristotle, Kant, Nietzsche. They all shaped me.

I still remember the first time I picked up a philosophy book at the library. I didn't even know what philosophy really was at the time, but I couldn't put it down. I was reading Plato's Republic, and there were many words I couldn't quite understand in English yet, but the whole thing fascinated me.

Even while building a career in technology, philosophy never left me. I kept questioning everything, searching for meaning and structure.

Over time, I realized that values are the starting point. That insight pushed me to reverse engineer my own thought process. Slowly, a pattern emerged. I later discovered that parts of this pattern had striking similarities to ideas developed by classical thinkers such as Aristotle, Saint Augustine, and Thomas Aquinas.

After I learned about Thomas Aquinas, he became my primary reference point for shaping the framework. His account of Intellect and Will closely matched what I had already mapped out. But I diverged from him in two important ways.

I placed Conscience after the Will, as the judge of an action, and I introduced Spirit as the integrator and guardian of the loop.

Spirit was the hardest concept for me to crystallize.

The word "Spirit" is loaded. It can evoke ghosts, supernatural ideas, and superstition. That's not what I mean by it.

For me, Spirit represents a feedback loop. Not feedback from a single action, but feedback from an entire trajectory. It is the ability to step back and look at the history of your actions and decisions, and to experience a sense of harmony or discontent with the direction of your life.

Later, I discovered that Thomas Aquinas had developed a related concept called "habit," building on Aristotle's virtue ethics. But habit, as Aquinas describes it, wasn't quite what I had in mind.

Habit is a stable disposition that shapes how we act. Spirit, in SAF, is dynamic. It is the culmination of the process that closes the loop and feeds the result back into the Intellect for the next iteration.

I developed this framework long before AI.

It was born out of pure curiosity and a desire to understand the human mind.

I also became Catholic along the way.

For a long time, I kept the framework mostly in my head. The first version I put on paper was a self help guide, which I've linked on the homepage.

Then ChatGPT came out.

I asked it to help me turn the framework into an ebook, and it was during those sessions with ChatGPT that the idea of building SAFi emerged.

One of the questions I had always asked myself was: Does the framework actually work? And how could I validate it?

The only way I knew to begin validating an idea scientifically was to formalize it and submit it for peer review in an academic journal.

And honestly, I assumed that as soon as reviewers saw words like "Spirit" and "Conscience," it would be an automatic rejection. 😁

But AI presented an interesting opportunity.

If an AI model could provide the reasoning component, I could actually implement the framework in software and observe how it behaved in practice.

That's what runs the demo today.

There is, however, a sensitive part of this entire project, and that is the danger of confusing the map for the terrain.

SAFi implements what I believe is a useful map of the human mind. But SAFi itself is not a human mind.

It borrows the structure of the framework and implements it in software, but it does not possess the human qualities that make a human being what they are.

Things such as free will, synderesis, conscience, and Spirit, as I understand them in human beings, cannot simply be recreated in software.

This isn't unusual in engineering.

We often borrow concepts from nature to build things that aren't nature. An airplane borrows language from birds to describe aspects of its aerodynamics. AI uses terms such as memory, reasoning, attention, and learning, all of which are deeply rooted in human experience.

SAFi does something similar.

It borrows concepts from classical philosophy to create a structured and auditable reasoning pipeline.

So I hope the philosophical terminology doesn't scare you off or make you uncomfortable.

I'm not claiming that SAFi has a human intellect, a human conscience, or a human spirit.

I'm saying that these concepts provide a remarkably rich and well tested conceptual framework for thinking about how values can guide reasoning, how reasoning can guide action, and how the consequences of those actions can feed back into the system.

The map isn't the terrain.

But sometimes a good map can help us understand the terrain a little better.

Nelson