Collaborators: Ron Rivest played an important early role in motivating the creation of the project, by pointing out to NIST that a public source of randomness could be valuable for auditing voting machines. Michael Fischer was a valuable early collaborator in thinking about a theoretical framework for public randomness. Andrew Regenscheid provided valuable administrative and technical support to the project. Overall, the NIST Beacon project has motivated several outputs, by the Information Technology Laboratory (ITL) and the Physics Measurement Laboratory (PML), involving collaboration from various NIST members/associates, including Michael Bartock, Lawrence Bassham, Joshua Bienfang, Peter Bierhorst, Harold Booth, Luís Brandão, Tyler Diamond, Thomas Gerrits, Scott Glancy, Michaela Iorga, John Kelsey, Emanuel Knill, Paulina Kuo, Alan Migdall, Carl Miller, Sae Woo Nam, René Peralta, Andrew Rukhin, Krister Shalm, Michael Wayne.

• 2011: Start of the Beacon project at NIST, to promote the development of randomness Beacons.

• 2012: Awarding of a multi-year grant from the Innovations in Measurement Science Program at NIST, for the Physics Measurement Laboratory (PML) to perform research on quantum random-number generation, and for the ITL to develop an implementation of the NIST Randomness Beacon.

• 2013: Deployment of the first public prototype of the NIST Beacon (version 1.0).

• 2015: Experimental validation of Bell loophole-free inequalities.

• 2018: Development (by PML) of a quantum random-number generator based on photon detection.

• 2018: Upgrade of the NIST Randomness Beacon to version 2.0.

• 2019: Publication of the "Reference for Randomness Beacons".

The current reference (2.0) for randomness beacons is the NIST Internal Report (NISTIR 8213) “A Reference for Randomness Beacons: Format and Protocol Version 2” (draft).

The publication is available free of charge from https://doi.org/10.6028/NIST.IR.8213-draft.

There was a period of public comments open in 2019. We plan to have the final version published in 2021. You may send comments to beacon-nistir@nist.gov.

We expect this reference document to promote the development of technology related to uses of public randomness for privacy-preserving auditability applications of societal benefit.

Some features of a beacon, as defined by the new reference:

Periodically pulsates randomness (e.g., once a minute). Each pulse has a fresh 512-bit random string, cryptographically combining entropy from at least two separate random number generators (RNGs). Each pulse is indexed, time-stamped and signed. Any past pulse is publicly accessible. The sequence of pulses forms a hash chain. Far-apart pulses can be efficiently verified via a short chain (skiplist). A pre-commitment of local randomness enables securely combining randomness from multiple beacons.

It is challenging to implement a secure and reliable randomness beacon matching the devised reference. Ensuring a reliable production of timely randomness requires understanding the functioning of the Beacon engine and its interface with associated machinery. Enabling the retrieval of any past pulse and associated information requires implementing and maintaining a high-availability public online interface.

The NIST Randomness Beacon is available online at https://beacon.nist.gov/home

The following image is a high-level depiction of some components of the Beacon service. The NIST Beacon integrates a PML-developed quantum-RNG (identified in the figure as #3) based on photon detection.

A Beacon implementation does not require knowing who are the actual users of the published randomness. The Beacon service only interacts with users via the public query-reply web-interface.

The Beacon engine has limited interactions accepting input from the outside:

• it interacts with a time server for the purpose of clock synchronization;
• it may use publicly verifiable "external values" for proving that some pulses could have not been pre-computed before certain time-marks.

List of Beacons in the process of implementing the new reference for randomness Beacons.

• United States: NIST Randomness Beacon

• Chile: CLCERT Randomness Beacon

• Brazil: Brazilian Randomness Beacon

We would like others to join … and we expect this list to continue growing.

We plan to develop guidance about the usage of Beacon-issued randomness. A few examples are given in the reference for randomness beacons.

Example applications of Beacon randomness:

• Select test and control groups for clinical trials.

• Select random government officials for financial audits.

• Assign court cases to random judges.

• Sample random lots for quality-measuring procedures.

• Provide entropy to digital lotteries.

Some generic goals:

• Enable public verifiability of random sampling.

• Prevent auditors from biasing selections (or being accused of it) and auditees from knowing the selections in advance.

The project is also interested in assisting complementary initiatives of research and development about trusted randomness, e.g., about quantum random-number generators and certifiable randomness.