How do you make mass out of nothing but waves?
That's the question the AI on the other page is trying to answer — and it's a 300-year-old question with a fresh twist. Part 1 below is the short history of asking what gravity actually is; Part 2 is a tour of what a wave on a field really looks like, because that's where the answer might live.
Scroll. The picture on the right will follow.
scroll to step through · part 1 is illustration · part 2 is interactive 3D
Newton said gravity is a force.
Before 1687, "gravity" was just the unexplained tendency of heavy things to fall. Everybody knew apples fell. Nobody had a maths for it.
Newton's wild move was to claim that the same equation that drops the apple keeps the moon in orbit. One law, whole universe. F = G·m₁m₂/r². For 200+ years it predicted everything you could see in the sky.
Newton himself wasn't comfortable. His force acts instantly across empty space — touch one mass, the other feels it at the same moment, no matter how far apart. He called this "absurd" and admitted in his own notes that he didn't know the why, only the what.
Then Einstein said it's not a force — it's the shape of space.
Two and a half centuries later Einstein offered a why. His move was wilder than Newton's: there is no gravitational force at all. Heavy things just bend the geometry of spacetime around them, and other things move along the bent shape.
Imagine pressing a bowling ball into a stretched bed sheet. The sheet curves. Roll a marble nearby and it spirals into the dip — not because the bowling ball pulled on it, but because the floor it was rolling on changed shape. That's gravity.
Wildly specific predictions, all confirmed: light bends around the sun, time runs slower in stronger gravity (your GPS literally has to correct for this every millisecond), black holes exist where spacetime is curled up so tightly that not even light climbs back out.
General relativity is the most experimentally confirmed theory we have. It just works.
Meanwhile, everything else became quantum. Except this.
While Einstein was finishing GR, a different revolution was happening at the small scale. Tiny things turned out to behave like waves and particles. Probabilities replaced certainties. Energy came in discrete chunks.
From this came quantum field theory: every particle is a ripple in a field, every force is the exchange of particles between charges. Three of the four fundamental forces — electromagnetism, the weak nuclear force, the strong nuclear force — all fit one mathematical framework, to ridiculous precision.
Gravity refused to fit. Try to make a quantum theory of it by imagining "gravitons" exchanged between masses and the calculations blow up to infinity. General relativity works at huge scales. Quantum field theory works at tiny scales. Nobody knows how to merge them.
Einstein spent the last 30 years of his life trying. He died in 1955 with notes still on his bedside table. 70 years later, we still don't have an answer.
What people are trying right now.
A handful of serious approaches, each with smart people defending it:
String theory — what if particles aren't dots but tiny vibrating strings? Different vibrations would be different particles, and gravity would come along for the ride. But it needs ten or eleven dimensions, and after forty years it hasn't made a testable prediction.
Loop quantum gravity — what if spacetime itself is made of tiny loops, woven together at the smallest scale? An atomic structure for space.
Causal sets — what if spacetime is a discrete network of events with "before" and "after" but no continuous geometry between them?
Emergent gravity — what if gravity isn't fundamental at all? What if it emerges from something simpler doing its thing — the way temperature emerges from molecules moving?
None has won. Each has open problems nobody's solved. We don't know which (if any) is right.
This project takes the emergence route.
If gravity isn't fundamental — if it's something that appears when simpler stuff does its thing — then the natural place to look is fields. Fields are how every other particle and force we know about is built.
Vibrating fields could in principle do all kinds of things — including produce stable structures, including produce forces between them. The question is whether one of those forces can be gravity: universal, always-attractive, falling off with distance — rather than something signed and phase-dependent like electromagnetism.
To follow what the AI is searching for, you need a feel for what a wave on a field actually is. So let's look at one.
A field is something that has a value at every point in space.
Temperature is a field. The temperature is some number — say 21°C — at every point in your room. If you wrote down the temperature at a million points, you'd have a snapshot of the field.
The simplest possible wavy field is one where the value oscillates up and down over time. Watch the dots: each one is a point in space, and its size is the value of the field there. Together they breathe — a wave, in 3D.
This is called a scalar field. "Scalar" just means "one number, not a direction".
Now imagine the field is an arrow, not just a number.
At every point in space, the arrow points somewhere and has some length. As the wave passes by, the arrow swings up and down on a fixed line — the red curve traces the swing.
This is called linear polarization. Light from your phone screen does exactly this. The wave moves along the cyan axis; the arrows swing across it.
The arrow and the direction the wave moves are perpendicular — and independent. That's a thing 2D pictures hide.
The arrow doesn't have to swing on a line. It can spin.
If two components of the arrow are a quarter-cycle out of phase, the tip traces a circle as the wave moves past. A helix threads down the axis.
This is literal spin. It's where the word comes from in physics. Particles that "have spin" are like little circularly- polarized waves: the field around them is always turning.
You can only do this with a vector field — a scalar field has no "direction" to turn.
What if the arrow points along the wave's direction?
Then there's no waving across — there's bunching and thinning. The dots crowd together, then spread apart, in the direction the wave is going.
This is sound. Air molecules don't swing sideways when sound passes through them — they compress and release along the direction of travel.
Compression waves are everywhere — sound, seismic P-waves, the way a Slinky transmits a push.
In real 3D, waves from a point make expanding shells.
Pictures of waves usually show rings on a pond — but that's a 2D cheat. In 3D, the surface of constant phase is a sphere. The wave goes outward in every direction at once.
Freeze a wave like this and confine it, and what you get is the shape of an atomic orbital — the standing-wave shells around an atom's nucleus.
Every atom you've ever touched is a standing version of this.
Now: what happens when two waves cross?
Watch the picture. Two scalar waves are running through the same region of space, at different angles. With ordinary physics, they just pass through each other — the bumps line up for a moment, then keep going.
But if the field has the right kind of self-attraction — if a tall lump of field pulls more field toward it — then the crossing doesn't pass through. It locks. A small region of the field stays still and breathes in place, even though the original waves are still trying to push through.
That breathing, stuck region has energy. It has a location. It pushes back if you try to move it. It can attract other lumps just like it.
It behaves exactly like a particle with mass.
The AI is hunting for recipes — combinations of self-attraction, coupling, and field structure — that make this lock happen and behave the way real gravity behaves: always pulling, always falling off as 1/distance², the same for every kind of wave.
watch the AI hunt live →Now you know what the seven questions on the homepage mean. Stays the same when waves flip · works for both charge signs · always pulls · fades like 1/r² · stronger when there's more energy · cares about energy, not type · the maths checks out. Each is a test the recipe has to pass.
Want the actual equations behind everything you just saw — the wave equation, the potentials, the Q-ball ansatz, the mediator action, the AI's fitness scoring? See the Maths page →
Want to play with the wave visualiser yourself? open the full version →