The next generation of scientific discovery won't come from AI replacing scientists. It will come from AI empowering scientists to ask better questions, solve more complex problems and accelerate discovery.
That's the vision behind the U.S. Department of Energy's inaugural Genesis Mission, and Old Dominion University has been selected for two Genesis Mission awards that position its researchers among those helping build what DOE calls "the world's most powerful integrated science discovery platform."
Working with the Department of Energy’s Thomas Jefferson National Accelerator Facility (Jefferson Lab), Pacific Northwest National Laboratory and other national laboratory partners, Old Dominion University researchers will develop complementary AI technologies that help scientific facilities operate more efficiently while ensuring those systems remain secure, reliable and trustworthy.
"This recognition reflects Old Dominion University's growing role in addressing some of the nation's most ambitious scientific challenges," said Dr. Kenneth J. Fridley, vice president for research and economic development. "Our researchers are helping shape the future of AI for scientific discovery while strengthening Virginia's position as a leader in research, innovation and economic growth."
Building Smarter Scientific Facilities
One Genesis Mission project, led by Dr. Monika Yadav, assistant professor of physics and faculty member in the School of Data Science, will develop an “AI-Enabled Digital Twin for Scalable Injector Optimization and Control.”
Modern scientific facilities rely on particle injectors that must be precisely tuned before experiments can begin. This process often requires experts to manually adjust thousands of interconnected parameters. Yadav’s team is developing an AI-enabled digital twin that combines physics-based simulations, operational data and expert knowledge to predict how these complex systems may respond before changes are applied to the actual equipment.
The system is intended to support scientists and accelerator operators by enabling faster, more informed and more reliable decisions. By reducing the time required to optimize accelerator performance, it could allow researchers to spend more time conducting experiments and advancing scientific discovery.
During Phase I, the team will develop and demonstrate AI-driven research workflows and evaluate whether these approaches can improve predictions, guide experimentation and accelerate discovery.
“AI has the potential to improve many aspects of accelerator science, from design and simulation to operation, control and maintenance,” Yadav said. “Our goal is to develop tools that work alongside scientists and operators, helping them understand complex systems, make informed decisions and use accelerator facilities more efficiently.”
Building More Trustworthy Scientific Facilities
A second Genesis Mission project, led by Dr. Jiang Li, professor in the School of Data Science and Department of Electrical and Computer Engineering, focuses on ensuring AI systems themselves remain secure and reliable.
The project, "Physics-Aware Guardrails for Real-Time Integrity Protection of AI-Driven Scientific Workflows,” will develop AI guardrails capable of detecting cyberattacks, data manipulation and other threats that could undermine scientific research. The project is a nine-month feasibility study led by Old Dominion University in collaboration with Jefferson Lab.
Rather than relying solely on conventional cybersecurity methods, the team will combine artificial intelligence with physics-based validation to distinguish legitimate scientific phenomena from malicious interference, helping researchers maintain confidence in AI-assisted decision-making while protecting the integrity of mission-critical scientific workflows. The proposal describes a "secure-by-design" framework to detect attacks, identify compromised AI models and recover physically consistent results.
"As AI becomes more deeply integrated into scientific research, scientists need to know they can trust the results it produces," Dr. Li said. "We're building security into the scientific workflow from the start, combining AI with the laws of physics to detect when something isn't right and help protect the integrity of the discoveries that follow."
Complementary Approaches to AI for Discovery
Although the projects address different challenges, together they advance a common vision for the future of scientific research.
Dr. Yadav's team is developing AI that helps scientific facilities become smarter by improving performance and accelerating experimentation.
Dr. Li's team is developing AI that makes those same systems more trustworthy by ensuring they remain secure, resilient and grounded in scientific principles.
Together, the projects bring together researchers from ODU's School of Data Science, Department of Physics, Department of Computer Science, Department of Electrical and Computer Engineering, and Center for Accelerator Science, demonstrating the university's strength in interdisciplinary research.
Building on Momentum
Old Dominion University’s role in the Genesis Mission extends beyond its two lead projects. Dr. Rui Ning, assistant professor in ODU’s School of Data Science, serves as the University’s project lead on STRATOS (Security and Trust Runtime Architecture for Time-Critical Operational Science), a Genesis Mission project led by the University of Hawaiʻi at Mānoa in partnership with ODU and Argonne National Laboratory.
STRATOS will develop a security architecture designed to detect and mitigate cyberattacks targeting AI-enabled scientific workflows in real time. Ning will lead ODU’s development of efficient detection methods and contribute to real-time mitigation, helping ensure that the next generation of AI for scientific discovery is not only more capable, but also secure and resilient.
Together, the three projects demonstrate Old Dominion University’s expanding role in advancing the performance, security and trustworthiness of AI for scientific discovery.