Probability Machine

Every quantum computing headline promises the same thing: a machine that will break encryption, design miracle drugs, and rewrite the limits of computation. Almost none of them mention the part that actually matters — a qubit's state collapses in a few hundred microseconds, wrecked by a stray photon or a defect no one can quite locate in the oxide.This book is reported from that second, quieter story.Its spine is the surface code and the brutal arithmetic of the error-correction threshold — the point where adding more physical qubits finally makes a logical qubit better instead of worse. On December 9, 2024, Google Quantum AI crossed that line with a 105-qubit chip called Willow, publishing an error-suppression factor of Λ = 2.14 in Nature. It was real. It was also roughly a thousandfold short of the machine the world has been promised.From there, the book travels to every architecture betting superconducting circuits are the wrong bet — trapped ions, neutral atoms, photons, silicon spins, topological qubits — and follows the money, the gamed benchmarks, and the cryptographic migration already underway.The final chapter names the four numbers that would settle the argument, and who is sitting on them.Read it before you believe the next press release.