Experimental Realization of Certified Randomness from Quantum Supremacy

PI Ruslan Shaydulin, JPMorgan Chase
Co-PI Matthew DeCross, Quantinuum
Michael Foss-Feig, Quantinuum
Minzhao Liu, JPMorgan Chase
Pradeep Niroula, JPMorgan Chase
Marco Pistoia, JPMorgan Chase
Yuri Alexeev, Argonne National Laboratory
Travis Humble, Oak Ridge National Laboratory
Shaydulin INCITE 2025
Project Description

While quantum computers are expected to solve a wide range of practically important problems beyond the capabilities of classical computers, realizing this potential remains a challenge. This project considers the problem of generating true randomness, which is crucial for applications ranging from cryptography to public lotteries. At this task, quantum computers have an advantage over classical computers due to their intrinsic probabilistic nature. A fundamental problem shared by known methods of generating randomness is that a client in need of entropy often must trust a third-party or a black-box provider, a problem rendered severe in communications over public network with malicious adversaries. This has led to the formulation of "certified randomness," randomness that a skeptical client can independently verify with high confidence, without the need to check or trust the internal workings of the generator. While progress in quantum hardware has enabled experimental certified randomness, the known protocols have severe limitations in terms of practicality. The goal of this project is to implement a novel, more practical protocol for generating certified randomness, with generation performed on a quantum computer and certification on a classical supercomputer accessed through INCITE. This project will take a step towards improving practicality of a useful task performed by a gate-based quantum computer in conjunction with a classical supercomputer.

Allocations