Controlling quantum devices isn't easy — you need to be an expert in both physics and programming. Researchers taught a neural network to tune superconducting qubits — the building blocks of quantum computers — on its own. Like an autopilot, it figures out the instruments and runs experiments without any hints. Imagine your smartphone optimizing its microchips while you sleep.
An AI has learned to conduct the quantum orchestra. Its instruments: superconducting circuits—artificial atoms kept colder than interstellar space. This AI-conductor designs and runs entire experimental scores, not just playing notes but creating them on the fly.
In one performance, it executed resonator characterization, mapping a component’s frequency response without guidance. It also recreated a quantum non-demolition measurement—like plucking a violin string so gently that other strings remain silent. Such finesse is critical because quantum systems easily lose their harmony, a process linked to entropy, the measure of disorder.
The result: quantum hardware becomes accessible. Researchers describe an experiment in plain language, and the AI translates it into precise commands. This builds on the superconductivity insights of John Bardeen and advances the vision of David DiVincenzo for scalable quantum computing.
🎯 These circuits are chilled to less than 0.1 degrees above absolute zero—colder than interstellar space—to keep their quantum performance undisturbed.
🎬 In the film 'Her', an AI assistant handles life’s chores with human-like intuition. This experimental framework offers a similar helping hand for quantum physicists in the lab.