1944: Ernő Rubik is born

On 13 July 1944, Ernő Rubik was born in Budapest, Hungary. The son of an aircraft engineer and a poet, he grew up to become a sculptor, architect and professor of architecture. His most famous creation began as a classroom aid for teaching three-dimensional geometry and design principles.

In the spring of 1974, while teaching at the Budapest College of Applied Arts, Rubik built a prototype of what he first called the "Magic Cube". The device comprised 26 smaller cubes that could rotate freely about a central core while still remaining attached — an elegant piece of mechanical engineering. After he patented the design and overcame years of production challenges, the cube reached the world market in the early 1980s and rapidly became the best-selling puzzle toy of all time.

Although Rubik never intended his invention as a contribution to computer science, the cube almost immediately captured the imagination of artificial-intelligence researchers. Solving it requires search through a staggeringly large space: more than 43 quintillion possible arrangements. Yet it is always possible to restore a scrambled cube to the solved state in at most 20 moves — a number known as God’s number. Early AI systems applied classic heuristic-search methods such as A* and IDA* to find short solutions. Later research turned to evolutionary computation, neural networks and, most recently, deep reinforcement learning. In 2019, researchers demonstrated a neural network that learns to solve randomly scrambled cubes without ever being shown the cube’s rules, simply by generating and practicing millions of its own solves.

The Rubik’s Cube therefore grew into a canonical testbed for planning, combinatorial search, group-theoretic techniques and end-to-end machine learning. Its lasting popularity has kept the computational frontiers of complex problem-solving visible and accessible to successive generations of AI researchers.

Today, while large language models and general agents confront ever richer sequential decision tasks, the brightly coloured plastic cube remains a miniature laboratory. New algorithms can still be tried, ranked and refined upon it — a reminder that inspiration for artificial intelligence sometimes arrives through an architect’s playful curiosity rather than a formal research agenda.