Unpredictable first, visible afterward
NIST’s June 2025 account describes CURBy, a University of Colorado Boulder and NIST service that publishes random bits derived from measurements of entangled photons. A Bell test checks correlations beyond what a classical local model allows. The aim is both an unpredictable source and a record that others can inspect. [1]
- Quantum measurements supply uncertainty
- Processing leaves a traceable record
- Published bits support an auditable choice
The path from measurement to published bits
The researchers’ November 2024 preprint separates the physical source from the processing that extracts usable randomness. Its protocol links records through intertwined hash chains—digital fingerprints connecting pieces of data—so the extraction process can be traced. The authors also state assumptions about measurement choices, timing and secure recording equipment; certification does not mean an assumption-free machine. [2]
Public is the feature—and the limit
NIST’s beacon guidance describes public randomness as useful for auditable randomized processes. It explicitly warns against using beacon output as secret cryptographic keys. A number that was difficult to predict before publication becomes visible afterward. A selection can benefit from that visibility; a secret cannot. [3]
Go a little deeper
Optional reading · about 1 more minute
A drawing that other people can check
Hypothetical example: A club publishes its entrant list, selection rule and chosen future beacon time before a drawing. Afterward, members can apply the same rule to the published value. This illustrates why agreement about the procedure matters alongside the source of randomness; it is not a complete lottery design.
What has not been checked here
Our interpretation: A historical experiment and its protocol are different evidence from today’s service availability. This explainer does not test current uptime or certify any particular drawing. The original paper and public-beacon guidance are linked so readers can distinguish the physics from the application.
Original sources
Attributed synthesis, not original reporting. Examples labeled hypothetical or illustrative are explanatory. Reviewing a source does not independently validate its findings.
- NIST: a factory for random numbers ↗
June 11, 2025, updated June 23; Bell-test mechanism and public purpose read September 26. Historical report, no current uptime claim.
- Kavuri and colleagues: traceable quantum randomness ↗
November 8, 2024 preprint; abstract and introduction, assumptions and extraction read September 26. Linked final Nature page did not yield usable text; preprint scope retained.
- NIST CSRC: interoperable randomness beacons ↗
Created June 3, 2019, updated February 5, 2025; public auditability and secret-key warning read September 26. Older roadmap dates not treated as current.
