ORNL researchers showcase AI-driven science at national expo

Events like the AI+ Expo are critical because they bring together varied stakeholders to align capabilities around a shared goal: dramatically increasing the productivity, impact and competitiveness of American science and technology.

Demonstrating 3D printing, quantum materials and autonomous laboratories

Other ORNL presenters included senior computational scientist Stephen DeWitt, who demonstrated an AI-driven system called LOOP that improves metal 3D printing by monitoring parts in real time and automatically adjusting the manufacturing process. The system combines AI agents, supercomputing and advanced manufacturing tools to produce reliable parts for critical energy infrastructure more efficiently.

Yongqiang Cheng, who serves as a distinguished staff scientist at ORNL's Spallation Neutron Source, also demonstrated how AI and high-performance computing can help scientists better understand quantum materials by uncovering hidden spin interactions that drive their behaviors. Along with his team, their research showed how AI and advanced computing methods can analyze experimental data faster and more accurately, helping accelerate the discovery of materials for future technologies such as advanced electronics and quantum computing.

Distinguished research scientist Rafael Ferreira da Silva demonstrated how ORNL uses OPAL, the Orchestrated Platform for Autonomous Laboratories, together with AI and supercomputing to automatically monitor and analyze plant growth at the Advanced Plant Phenotyping Laboratory. The system combines several types of plant images and processes them on the Frontier supercomputer to quickly predict plant traits across large populations, collapsing week-long detection cycles into sub-minute autonomous responses.

"At ORNL's Advanced Plant Phenotyping Laboratory, we are building a specialized set of AI agents to coordinate imaging instruments, exascale inference on Frontier, and experimental decision-making inside a single closed loop," said Ferreira da Silva. "We're moving past the era of AI as a research assistant and into one where agents actively run the science. Coupling foundation vision models with exascale-class agentic workflows turns high-throughput plant phenotyping into a discovery engine that learns continuously and steers itself, with scientists on the loop where it matters most."

Throughout the expo, ORNL researchers emphasized the laboratory's broader vision for converged computing, integrating AI, high-performance computing, quantum information science and autonomous scientific workflows. The discussions reflected ORNL's strategic focus on advancing energy-efficient AI, scientific automation and next-generation computing capabilities that strengthen U.S. leadership in science and technology.

UT-Battelle manages ORNL for the Department of Energy's Office of Science, the single largest supporter of basic research in the physical sciences in the United States. The Office of Science is working to address some of the most pressing challenges of our time. For more information, please visit energy.gov/science. - Mark Alewine