Preface
The world is full of motion, light, sound, heat, and invisible forces—and in your pocket is a scientific instrument powerful enough to explore them all. This book is about curiosity, creativity, and using everyday technology to uncover the hidden rules of the universe. Whether you’re a student, a teacher, or an independent learner, the following pages will help you turn your smartphone into a physics lab.
Before we dive into the experiments, take a moment to read the next few sections. They’ll introduce the project, how to get the most out of it, and how to stay safe while exploring.
A Word of Caution
This book you are holding is itself an experiment. It was created not in the traditional way, but in collaboration with an AI — specifically, ChatGPT. That doesn’t mean I simply asked, “Hey ChatGPT, write a book on physics experiments with AI” and waited for magic. Far from it.
Instead, it began with about forty pages of rough notes — a collection of physics experiments you could do with a smartphone. These notes started to take shape while I was writing a book on Android programming, where I became amazed by the sheer number and power of the sensors packed into modern smartphones. But the real spark came when we got the CAT S60, the so-called “Thermo-Smartphone.” I was instantly hooked: first, when I pointed its thermal camera at a mixture of ice and salt and immediately saw a vivid reading of –16.2 degrees Celsius; and second, when I used its high-speed magnetometer (sampling at 100 Hz) to capture the changing magnetic fields from a running hair dryer. Right there, visible in the raw data, was the 50 Hz signal — and with it, the beautiful dance of Shannon-Nyquist sampling theory made real. Add to that a stunning lunar eclipse that same year, and from then on, I found myself scribbling down every clever experiment I came across, especially those seen on YouTube.
Those notes were originally just for myself. I never really intended to do much with them, especially because so many excellent resources already existed. Foremost among them is phyphox from RWTH Aachen University — a brilliantly designed platform offering a wide range of smartphone-based experiments, complete with tailored apps and clear documentation. Then there’s Practical Physics, a project of the Nuffield Foundation and the Institute of Physics, packed with fantastic, classroom-tested ideas. Other inspiring sites include Physics Open Lab, with its beautifully documented advanced setups; PhysicsExperiments.eu, a treasure trove of simple yet profound demonstrations; Science in School, connecting science education with real-world research; and The Naked Scientists, whose lively experiments bring science to life. With so much great material already available, I wondered: what difference could my little set of notes possibly make?
Adding to that, I have a full-time job teaching computer science, and these days, physics has become more of a passionate hobby than a profession. Realistically, there was no way I could find the time to write the kind of book I envisioned. And even though I am a theoretical physicist by training, I strongly believe that physics — especially in a school context — should be taught through experiments first, not through reliance on mathematics. In my experience, it’s often the early and overwhelming focus on math that discourages young people from pursuing careers in science or engineering. Physics should be about curiosity, discovery, and wonder first; the mathematics can come later, when the excitement is already there.
Then came AI — or more specifically, large language models like ChatGPT from OpenAI, Gemini from Google, and others. Like many, I was first surprised, then amazed, and naturally scared by the capabilities of these systems. As I started experimenting, I luckily had the idea to use my old smartphone physics notes as a playground. Initially, my goal was simple: could ChatGPT help me with the math? As it turns out, not only could it assist with calculations, but it could do much more — acting as a personal tutor, providing context, suggesting ideas, and even serving as a sparring partner for exploration. It can offer the kind of individualized support that a single teacher would love to give every student — but realistically cannot provide when teaching 20 or 30 at once.
Once it became clear that my original idea of serious physics experiments without math had evolved into physics experiments with AI, a new question arose: how could I possibly write this book given my time constraints? In computer science, there’s a saying: “Eat your own dog food” — meaning, if you believe in your own tools, you should use them yourself. So I thought: if this is a book about using AI, why not write the book with AI?
This was a three-week-long, highly interactive process — shaping, questioning, correcting, and constantly refining ideas together with ChatGPT. The result is not a perfect book; it’s a living document, created in a world where AI, science, and education are evolving faster than traditional publishing can keep up. I chose to embrace that speed, to move quickly, and to prioritize inspiration over polish. This book is meant to be a spark — a starting point for you to continue with your own ideas.
A short note on the experiments themselves: quite a few of them I performed personally, so I know they work. Many were inspired by the websites mentioned earlier, and I always tried to credit the original sources where possible. Some ideas — especially the really challenging or striking ones — came from YouTube, and while not all of them involve smartphones directly, they were simply too fascinating to leave out. A few experiments were suggested by ChatGPT itself. They seemed both doable and interesting, so I included them as well.
Particularly the experiments in Part 4 grew out of a series of conversations with ChatGPT. Originally, I was just fooling around, not expecting much to come from it. But surprisingly, ChatGPT showed a real enthusiasm for biology-related experiments — quite curious, considering it is not biological itself (!?!). I think these experiments are neat, and they may be especially interesting for teachers working in these fields. They make no real claim to originality, but I believe they are worth mentioning nonetheless. That is also why, in this part, there are basically no references listed.
In closing, I would like to mention the project’s website: sphysix.org. The idea behind it is simple — to provide all the material from this book under an open license, free for everyone to use, adapt, and share. My hope is that it might grow into a small community, a place where we can exchange experiences, correct errors, share new discoveries, and spark fresh ideas together. I would truly love to hear from you. Welcome to Smartphone Physics Experiments with AI — and happy exploring!