SIGNALprophet format
Quantum advantage has always hinged on the query ledger — how many times a black box gets touched, and what counts as a touch. A new arXiv Comment on a 2026 thermodynamic result says the original ledger doesn't balance, and the failure is mechanical, not arithmetic.
The 2026 paper asked whether a single controlled temperature change to a probe could decide the Deutsch-Jozsa problem — a foundational quantum benchmark that asks whether a hidden function is constant or balanced — and argued the rest of the work is done by roughly 116 cheap ancillary measurements. The Comment's load-bearing point is sharper than "the math is off": when the original protocol used a CNOT fanout to spread the probe's readout across many auxiliary qubits, those qubits end up perfectly correlated, not independent. They are not 116 samples of the same bit. They are 116 copies of one bit, each carrying the same value — which is to say, no new information per extra measurement.
This is the kind of step that looks like amplification and functions as a wire. Independent samples would instead require preparing and cooling fresh probes each time, which under the original paper's own definition counts as repeated queries — collapsing the "one thermal query" headline back into many.
The Comment does not reject the underlying physics. The thermal-kickback idea — that the probe's temperature itself can still carry the answer — survives the narrow, and is the line the original authors have to answer next.