Building a Quantum Random Number Generator

A classical "random" number generator is actually pseudo-random: a deterministic algorithm that just looks random, given the same seed, it always produces the same sequence. A qubit in superposition is different, its measurement outcome is fundamentally, physically unpredictable, not just hard to predict. This lesson builds a genuine quantum random number generator (QRNG).

The idea

Put every qubit into an equal superposition with h, measure them all, and read the resulting bitstring as a binary number:

from qiskit import QuantumCircuit
from qiskit_aer import AerSimulator

def quantum_random_bits(n_bits):
    qc = QuantumCircuit(n_bits, n_bits)
    qc.h(range(n_bits))              # put every qubit into superposition
    qc.measure(range(n_bits), range(n_bits))

    simulator = AerSimulator()
    result = simulator.run(qc, shots=1).result()  # one measurement, one random bitstring
    bitstring = list(result.get_counts().keys())[0]
    return int(bitstring, 2)         # parse the bitstring as base-2

print(quantum_random_bits(8))   # a random integer from 0 to 255
print(quantum_random_bits(8))   # a different one, genuinely unpredictable

• qc.h(range(n_bits)) applies h to every qubit at once, range(n_bits) expands to [0, 1, ..., n_bits - 1], and Qiskit accepts a list of qubit indices anywhere a single index is accepted.
• shots=1 is deliberate here, unlike the last lesson, this isn't about estimating a probability distribution, it's about getting exactly one genuinely random outcome.
• int(bitstring, 2) parses the measured bitstring (like '10110100') as a base-2 number, converting 8 random bits into a random integer from 0-255.

Why this matters

Real quantum random number generators are used in cryptography today, precisely because their unpredictability doesn't rely on hiding an algorithm or a seed, it's a fundamental physical property. A classical pseudo-random generator, no matter how sophisticated, is deterministic under the hood, if you know the seed and the algorithm, you can predict every "random" number it will ever produce.