Quantum Measurements Cannot be Proved to be Random

We show that it is impossible to prove that the outcome of a quantum measurement is random.

💡 Research Summary

The paper tackles a foundational question at the intersection of quantum physics, information theory, and mathematical logic: can we ever prove that the outcomes of quantum measurements are truly random? The authors begin by distinguishing between two common notions of randomness. The statistical view treats randomness as a property that can be inferred from empirical frequency distributions and passed through a battery of tests (e.g., NIST, Diehard, TestU01). The algorithmic view, rooted in Kolmogorov complexity, defines a string as random if no program substantially shorter than the string itself can generate it. This second definition is the one the paper adopts because it is amenable to formal proof‑theoretic analysis.

Having fixed the definition, the authors review the relevant incompleteness results. Gödel’s first incompleteness theorem shows that any sufficiently expressive, consistent formal system cannot prove all true arithmetic statements. Chaitin’s information‑theoretic incompleteness theorem strengthens this by constructing a real number Ω (the halting probability of a universal Turing machine) whose binary expansion is algorithmically random. Crucially, for any given formal system S, there exists a bound N(S) such that the first N(S) bits of Ω are undecidable within S; no finite proof in S can determine their values. In other words, the statement “the k‑th bit of Ω equals 0” is true for some k but unprovable in S once k exceeds the bound.

The core argument of the paper maps this logical structure onto quantum measurement. A quantum random number generator (QRNG) repeatedly measures a qubit prepared in a superposition, producing a binary sequence. If we assume that the underlying physical process is ideal, the resulting sequence should be indistinguishable from a sequence drawn from a truly random source. The authors argue that to prove the sequence is random in the algorithmic sense, one would need to demonstrate that no shorter description exists—a task equivalent to proving that the sequence matches the bits of some Ω‑like number. Because of Chaitin’s theorem, such a proof cannot be carried out in any recursively enumerable axiomatic system that is capable of arithmetic. Hence the claim “quantum measurement outcomes are random” is a true statement that lies outside the provable realm of standard mathematics.

The paper does not stop at the abstract impossibility result. It surveys practical randomness tests used in quantum cryptography and QRNG certification, emphasizing that these tests are statistical hypothesis checks rather than proofs of algorithmic randomness. A sequence can pass all known tests and still be algorithmically compressible; conversely, a truly random sequence may fail a particular test due to finite‑sample fluctuations. Therefore, empirical validation can only provide confidence, not certainty.

In the final sections, the authors discuss implications for quantum cryptography, especially device‑independent quantum key distribution (DI‑QKD). In DI‑QKD, security is based on the violation of Bell inequalities rather than on an explicit proof of randomness. This shift reflects the recognition that randomness cannot be mathematically certified; instead, one must rely on physical assumptions (e.g., no‑signalling, isolation of devices) and on rigorous statistical monitoring.

The conclusion reiterates the main thesis: while quantum mechanics predicts probabilistic outcomes and experimental data strongly support this, the formal proof of randomness for any concrete measurement record is impossible due to deep logical limits. The authors advocate a pragmatic stance: focus on robust device design, continuous statistical testing, and security proofs that treat randomness as an operational assumption rather than a mathematically provable fact. This perspective aligns with current practice in quantum information science and highlights a subtle but important boundary between what physics can assert and what mathematics can prove.