DOE Selects CFS to Lead AI Digital Twin Project for SPARC

Commonwealth Fusion Systems announced on 8 ottobre 2026, that the U.S. Department of Energy selected it to lead a project in the second phase of the Genesis Mission, making CFS the only private company leading a Phase II award under the national AI-for-science initiative. The CFS-led effort, listed by DOE under the title “Accelerating the path to commercial fusion energy via an AI-enabled digital twin platform for SPARC,” is the only fusion-focused project among the new Phase II selections.

The Energy Department’s 8 ottobre 2026 announcement covered 12 Phase II Genesis Mission project awards totaling $159 million, issued through the department’s Office of Science. DOE did not break out funding amounts for the individual projects.

According to CFS, the project supports the development of an AI-enabled digital twin platform intended to help optimize operations of SPARC, the prototype fusion machine the company is assembling in Devens, Massachusetts. CFS states that SPARC is expected to achieve the first commercially relevant demonstration of net fusion energy and to pave the way for ARC commercial fusion power plants, starting with the Fall Line Fusion Power Station in Chesterfield County, Virginia.

“This project will convert decades of R&D performed across the U.S. fusion community into tools that can be applied to accelerate SPARC operations,” said Brandon Sorbom, Chief Science Officer and co-founder of CFS. He said the assembled team of public and private leaders in AI and fusion would work to unify and apply decades of U.S. research against some of the fusion industry’s biggest challenges.

Technical Scope and Project Team

The project will use DOE supercomputers to process decades of fusion data, strengthening AI tools and expanding on digital twin visualization groundwork started by CFS, NVIDIA, and Siemens. Per the company’s description, planned workstreams include building the software infrastructure to connect to DOE computing resources; building the training datasets and AI surrogate models needed for an AI-enabled digital twin; training reinforcement learning and agentic AI models to be used by an AI-based SPARC operations co-pilot; and building data visualization pipelines that accelerate operations. CFS states the resulting digital twin will help accelerate commercial fusion deployment by allowing rapid iteration in the SPARC control room.

“AI models trained on DOE supercomputers will be used to plan pulses in the SPARC control room and to rapidly make decisions based on experimental data as it comes off of the machine,” said Tom Looby, the CFS Principal Investigator leading the project. “AI is serving as a high-speed co-pilot, turning hours of heavy simulations into instant insights between plasma pulses.”

CFS’s team of 10 partners includes Lawrence Livermore National Laboratory, Oak Ridge National Laboratory, Princeton Plasma Physics Laboratory, and SLAC National Accelerator Laboratory; Carnegie Mellon University and the Massachusetts Institute of Technology; and Google DeepMind, NVIDIA, Synopsys, and General Atomics. The company characterized the project as the biggest collaboration to date on the application of AI in fusion, and cited its annual collaboration workshops, which have included digital twin and AI sessions, as part of its track record of organizing the scientific community around commercialization goals.

Genesis Mission Portfolio

President Trump launched the Genesis Mission in novembre 2025 to accelerate discovery and strengthen U.S. leadership in science, medicine, and technology, according to DOE. The department describes it as a national initiative to advance AI for science and technology, built around the American Science and Security Platform, which connects supercomputers, experimental facilities, AI systems, and unique datasets. DOE states that the goal of the Phase II awards is to scale up and expand the impact of projects that have shown potential for AI advantage and demonstrated a trajectory toward a transformative scientific capability; the selected projects are multiyear and use interdisciplinary teams.

DOE Under Secretary for Science Dr. Darío Gil, who serves as the department’s director for the Genesis Mission according to DOE’s program page, said the Phase II teams have already demonstrated what is possible when Super Intelligence is integrated with scientific expertise, data, and computing. He said the awards give the teams the opportunity to take that work to the next level.

The 12 new selections bring the total number of Phase II projects to 14, alongside previously announced awards for GridFM 2.0 under the Office of Electricity and Prometheus under the Office of Nuclear Energy. DOE is also issuing six new Phase I awards, capping a first-year cohort of 297 projects across the Genesis Mission portfolio.

The other newly awarded Phase II projects are led by the University of California San Diego, MIT, the University of California Irvine, the University of Illinois Urbana-Champaign, the University of Washington Seattle, Fermilab, Lawrence Berkeley National Laboratory, Harvard University, Argonne National Laboratory, Oak Ridge National Laboratory, and Northeastern University. Their stated focuses include expanding an RNA structure database for training SI models, mapping and managing deep geothermal reservoirs, designing microchips for extreme environments, and developing an automated assistant for the particle separator at the Facility for Rare Isotope Beams.

CFS describes itself as the world’s largest private fusion company and states that it has raised more than $4 billion in capital since its founding in 2018.