Langton's Ant: Two Rules, One Ant, and an Emergent Highway

Two rules, an ant, and a grid of squares

Langton's Ant, introduced by computer scientist Chris Langton in 1986, is one of the simplest systems in computer science ever shown to produce genuinely complex, unpredictable behaviour. The setup: an infinite grid of square cells, each either black or white; a single "ant" sitting on one cell, facing one of four directions. Every step, the ant follows exactly two rules, based only on the colour of the cell it is currently standing on:

on a WHITE cell: turn 90° clockwise, flip the cell to BLACK, move forward one cell
on a BLACK cell: turn 90° counter-clockwise, flip the cell to WHITE, move forward one cell

This is formally a two-dimensional Turing machine: the ant is a read/write head, the grid is an infinite tape extended into two dimensions, and the two rules are its entire program. There is nothing hidden and nothing random — the ant's future is completely determined by its starting position, orientation, and the two rules above.

Three phases: chaos, chaos, then a highway

Starting from an entirely white grid, the ant's path goes through three visually distinct phases. For roughly the first 500 steps, it draws small, simple, often symmetric patterns, revisiting cells near its start. From about step 500 to roughly 10,000, the pattern looks genuinely chaotic — dense, irregular scribbles with no obvious structure, the kind of trajectory that looks like it could continue forever without ever settling down.

Then, without warning, around step 10,000 (the exact number varies slightly by starting orientation), the ant escapes the chaotic mess and begins building a "highway": a diagonal strip of 104 steps that repeats forever, carrying the ant steadily off to infinity in a straight overall direction while tracing the same small repeating figure. Nobody has proven mathematically why the highway must eventually appear from every starting configuration on an initially blank grid — it is an empirical fact, observed in every simulation ever run, but the general question of whether Langton's Ant always eventually builds a highway from any finite initial black-and-white pattern remains an open problem in the mathematics of cellular automata.

Why such a simple rule can surprise you

Langton's Ant belongs to the same family of ideas as Wolfram's elementary cellular automata and Conway's Game of Life: minimal local rules, applied deterministically and repeatedly, can generate behaviour far richer than the rule itself suggests — a theme sometimes called emergence. It is also a favourite demonstration of why simulating ahead is often the only way to know what a simple deterministic system will do: there is no shortcut formula that tells you where the ant will be at step one million without actually running all one million steps, a property closely related to the idea of computational irreducibility.

Variants extend the idea with more than two colours and more than two turning rules — some of these produce even more elaborate structures, including some capable of the same kind of universal computation found in Conway's Game of Life.

Frequently asked questions

Try it live

Everything above runs in your browser — open Langton's Ant and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.