Researchers used the unique capabilities of a quantum computer to generate a verifiably random number in a study published in the journal Nature.
Even a workflow that included the world’s first exascale supercomputer couldn’t match the random quality of the quantum computer’s output.
“All of us in today’s world use randomness every day, whether we know it or not,” said Marco Pistoia, head of global technology applied research at JPMorganChase and principal investigator of the study. “Our communications on the internet — such as our banking transactions and the submission of our credentials via user IDs and passwords — all rely on encryption, which in turn requires the use of random numbers to generate encryption keys. This study brings us closer to the usability of quantum computing for practical, real-world applications and will be vital to further research in statistical sampling, cryptography and numerical simulations and other fields.”
Pistoia’s team used time on Quantinuum’s H2 quantum computer to execute random circuit sampling and perform a certified randomness expansion protocol, which outputs more randomness than it takes as input.
“These pioneering efforts push the frontiers of computing and provide valuable insights into the intersection of quantum computing and high-performance computing,” said Travis Humble, a coauthor of the study and director of the Quantum Science Center, a national quantum information science research center at the Department of Energy’s Oak Ridge National Laboratory.
The team verified the results by using a workflow that combined four supercomputers: Frontier and Summit at ORNL’s Oak Ridge Leadership Computing Facility (OLCF), Polaris at the Argonne Leadership Computing Facility (ALCF) and Perlmutter at the National Energy Research Scientific Computing Center (NERSC).
Researchers used the unique capabilities of a quantum computer to generate a verifiably random number, a potential step toward quantum advancements in encryption, network security and similar realms. The team cross-checked the results using classical supercomputers, including ORNL’s Frontier and Summit. Credit: Carlos Jones/ORNL, U.S. Dept. of Energy
“Even with this combined power, the classical machines failed to reach the same result in a comparable period of time,” Pistoia said.
Quantum computing relies on quantum bits, or qubits, to store information. Qubits, unlike the binary bits used in classical computing, can exist in more than one state simultaneously via quantum superposition, a quantum mechanical effect that allows combinations of physical values to be encoded on a single object. That dynamic enables a wider range of possible values, more like those of a dial with a range of precise settings than a binary on-off switch.
Researchers have theorized the expanded computational range enabled by quantum technology could provide new and more efficient ways to solve complex problems, from high-resolution digital simulations to unhackable encryption standards.
“The finance industry is also expected to benefit from quantum computing, and this is why JPMorganChase has invested in this technology,” Pistoia said. “So far, we’ve seen the current generation of quantum computers mostly used to solve ‘toy’ problems that can’t be applied at a practical scale because the computers aren’t powerful enough, or we’ve seen claims of quantum supremacy for contrived algorithms with no real application. This new result demonstrates a potentially more useful application.”
The research team used the Quantinuum computer, which employs trapped ions as qubits and offers all-to-all connectivity. This connectivity allows any qubit to easily connect with others. The higher the connectivity, the easier to entangle qubits — a necessary step for creating the quantum circuits that can run an application.
Pistoia’s team used a classical algorithm to randomly generate quantum circuits and submitted those circuits to the Quantinuum computer as inputs.
To certify the results, the team needed enormous computing power — more than any single supercomputer could offer. The team obtained compute time from a Director’s Discretionary allocation on the OLCF’s Frontier, the world’s first exascale supercomputer with speeds of up to 1.35 exaflops, or 1.35 quintillion calculations per second. The team also used the now-decommissioned Summit supercomputer, a 200-petaflop leadership-class machine at the OLCF; the 44-petaflop Polaris machine at the ALCF; and the 79-petaflop Perlmutter machine at NERSC.
The team combined the four machines in a workflow that fell short of the Quantinuum results.
“Classically, it would have taken more than 100 seconds to produce a similar number to what the quantum computer produced in 2 seconds,” Pistoia said. “This showed the results could not be mimicked by classical computing methods, as an attempt to mimic the result would cause a delay that would have been easily detectable.”
Cryptography standards rank randomness by bits of entropy, or the number of calculations or guesses necessary to crack a randomly generated passcode. The more variations, the more bits of entropy. A known password, for example, offers zero bits of entropy. The research team generated and certified 71,313 bits of entropy in their experiment.
Next steps include improving the certified randomness expansion protocol to make it useful for a wider range of applications.
Along with Pistoia and Humble, the research team included Minzhao Liu, Ruslan Shaydulin, Pradeep Niroula, Wen Yu Kon, Enrique Cervero-Martin, Kaushik Chakraborty, Omar Amer, Atithi Acharya, Shouvanik Chakrabarti, Danylo Lykov, Shaltiel Eloul, Niraj Kumar and Charles Lim of JPMorganChase; Matthew DeCross, K. Jordan Berg, Joan M. Dreiling, Neal Erickson, Cameron Foltz, Michael Foss-Feig, David Hayes, Michael Mills, Steven A. Moses, Brian Neyenhuis, Peter Siegfried, James Walker and Florian J. Curchod of Quantinuum; Yuri Alexeev and Jeffrey Larson of Argonne National Laboratory; and Shih-Huan Hung and Scott Aaronson of the University of Texas at Austin.
Support for this research came from the DOE Office of Science’s Advanced Scientific Computing Research program. The OLCF, ALCF and NERSC are DOE Office of Science user facilities.
Related news and publications
Liu, M., Shaydulin, R., Niroula, P. et al. Certified randomness using a trapped-ion quantum processor. Nature 640, 343–348 (2025). https://doi.org/10.1038/s41586-025-08737-1
Read more: JPMorgan Chase News Release | University of Texas News Release
ORNL continues to empower the pursuit of quantum innovation, advancing world-leading scientific discovery to enable a quantum revolution that promises to transform a vast range of technologies critical to American competitiveness. These traits are embodied by ORNL’s celebration of the International Year of Quantum Science and Technology in 2025. Click here to learn more about quantum science at ORNL.
UT-Battelle manages ORNL for DOE’s Office of Science, the single largest supporter of basic research in the physical sciences in the United States. DOE’s Office of Science is working to address some of the most pressing challenges of our time. For more information, visit energy.gov/science.
