Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Quantum Materials Ai topic
No spam. Unsubscribe anytime.
Presenters outline 'Genesis' AI mission to accelerate quantum materials research
Summary
Presenters described the Genesis mission94which leverages AI, supercomputers and instrument data94to speed discovery in quantum materials. They highlighted MakeMag for 2D quantum magnets, use of the Frontier supercomputer, and plans to share data via the American Science Cloud.
Presenters described the "Genesis" mission as a coordinated effort to speed discovery of quantum materials by combining artificial intelligence, high-performance computing and measurements from national research instruments. One presenter said the project aims to remove bottlenecks in understanding quantum materials that underpin emerging technologies such as quantum computing.
The presenters said Genesis will pair multimodal experimental data with digital twins and AI models. "Genesis mission is trying to explore and define a new way, of conducting scientific research by leveraging AI, supercomputers, and unique scientific instruments at DOE research facilities, we can really accelerate scientific discovery," Presenter 2 said. They identified MakeMag94"multimodal AI for 2D quantum magnets"94as a concrete example of that approach.
Presenter 4 described how neutron scattering and other probes feed the effort: "Neutron scattering is a technique where we take intense beams of neutrons and we scatter them off the materials, and that gives us fundamental information on the materials themselves and their fundamental properties, both their structure and dynamics, which can help us to make better materials for different applications." The presenters said combining data from neutron sources, light sources, nano centers and bulk characterization probes is intended to improve model accuracy for quantum magnets.
The team described user-facing applications for the tools. Presenter 4 said experiment users could interact with AI agents in natural language to describe how to run an experiment, lowering the barrier for novice users while helping experienced researchers analyze results faster. "It helps the novice user or somebody who's really good at discovering new materials, but may not be so expert in analyzing them, get results out faster as well," Presenter 3 said.
On the computing side, presenters said they run digital twins and train models on large supercomputers. "This supercomputer enabled AI driven workflow reduces the time from experimental solution from many months to just a few days," Presenter 2 said. The presenters also said data and trained models will be accessible through the American Science Cloud so that other researchers can reuse the resources.
Presenters emphasized the collaborative nature of the work and that the project has grown into a large team across institutions. They described Genesis and MakeMag as tools to accelerate materials research rather than as completed products; no formal decisions, procurement outcomes or policy changes were recorded in the session.
The presenters said the next steps include continued model training, further instrument integration, and sharing data and models through national cloud resources to support broader research use.
