What would it actually mean to solve physics?

We talk about finding the fundamental laws of nature, but as researchers we often don’t spend too much time thinking about what that actually means: whether a single set of rules can pin down our universe, to what extent experiments can distinguish that this is the case, and what can or cannot be computed in terms of such a fundamental description.

As AI starts to take on more theoretical work, the question feels newly practical: to automate physics, you’d first have to pin down exactly what the goal is.

This site is my attempt to teach myself enough background from foundational mathematics, machine learning, and metaphysics to understand this foundation better. I’ll keep track of my progress in a set of living lecture notes, and logbook posts about projects where I try to bring these ideas to life.

— Sabrina