Noise‑Driven 'Quantum Advantage' Claims

A bank–IBM paper reports a 34% gain in bond‑trade fill predictions after a 'quantum' data transform, yet the gain vanishes when the same transform is simulated without hardware noise. Aaronson contends the effect is a noise artifact and a product of unprincipled method comparisons and selection bias. He urges a proof‑before‑application standard: show real quantum advantage on benchmarks before touting finance wins. — It challenges corporate and media quantum hype and proposes a practical rule to prevent pseudo‑results from steering investment and policy.