Superposition

An ordinary light switch knows exactly two states: on or off. A quantum particle — an electron, for example — isn’t so clear-cut. It can exist in a state that contains both “on” and “off” at the same time. This state is called superposition. That sounds contradictory, but it isn’t a logical error: it’s a measurable property of nature. As soon as you measure the particle, it settles into a single state — but until then, it is literally both at once.

Superposition as the foundation of the quantum computer

Classical computers compute with bits. A bit is either 0 or 1, never both. Quantum computers work with qubits — these are quantum bits that, thanks to superposition, can be 0 and 1 at the same time. That sounds like a small difference, but it has enormous consequences.

With two classical bits, you can store exactly one of four possible values: 00, 01, 10, or 11. Two qubits in superposition represent all four values simultaneously. This advantage doubles with every additional qubit. A quantum computer with 300 qubits could theoretically represent more states simultaneously than there are atoms in the visible universe. That doesn’t mean it spits out all the answers at once — but it can dramatically shorten certain computational paths.

On its own, superposition isn’t enough. It’s the first step: it opens up the space of possible states. Other quantum effects — above all interference, meaning the targeted amplification of correct and suppression of incorrect computational paths — then exploit this space. Without superposition, none of these effects would exist. It is thus the foundation on which quantum algorithms are built.

How a qubit is brought into superposition

A qubit can be physically realized in various ways: as the orientation of an electron, as the energy level of an atom, or as the oscillation direction of a light particle (photon). The starting state is always well-defined — either 0 or 1. To create superposition, a so-called Hadamard operation is applied, a mathematically precise computational operation that transforms the qubit’s state into an exact mixture of 0 and 1.

The result is not a random state but a precisely defined one. The probabilities of obtaining 0 or 1 upon measurement can be calculated and controlled. This is exactly the difference from mere noise or randomness: superposition is controlled indeterminacy. The quantum computer uses this to perform computations in parallel within this superposed state, before a measurement finally fixes the result.

The biggest technical problem here: superposition is extremely fragile. Even the slightest disturbances from heat, radiation, or electromagnetic fields can destroy the quantum state — an effect called decoherence. That’s why quantum computers operate near absolute zero, around minus 273 degrees Celsius, far colder than outer space.

Superposition in news and products

The term regularly comes up when companies like IBM, Google, or the start-up IonQ unveil new quantum computers. In 2019, Google announced that its quantum chip “Sycamore” had solved a task in 200 seconds that would have taken a classical supercomputer 10,000 years. This was made possible by superposition and the quantum algorithms built on top of it.

Superposition also plays a role in cryptography. Certain quantum algorithms — the best known being Shor’s algorithm — could crack the encryption that currently protects bank data and passwords. This is precisely why security agencies worldwide are already working on new encryption methods that even quantum computers cannot break. When this becomes a real problem is disputed — but superposition is the reason this discussion is being had at all.

A common misconception: superposition does not mean that a quantum computer simply calculates faster than a normal one. It calculates differently. For most everyday tasks — writing text, playing videos, calculating spreadsheets — a classical computer is better suited. Quantum computers only offer advantages for very specific problems where the structure of superposition can be exploited: for instance, in simulating chemical molecules or searching through vast amounts of data.