The U.S. Department of Energy has selected Brookhaven National Laboratory to lead a $14.2 million project to build a next-generation AI system for the electric grid. The system, called a grid foundation model (GridFM), will simulate scenarios for adding new loads - the goal is one billion simulations in 24 hours - to speed up grid expansion planning.
The funding was announced September 1 at Brookhaven Lab by Catherine Jereza, DOE's Assistant Secretary for the Office of Electricity, which is administering the award. The project runs three years. Brookhaven Lab receives $3.9 million of the total, with the remainder shared among collaborating institutions. This is the eighth Genesis Mission award tied to Stony Brook University.
Who is building it
Hendrik Hamann, Brookhaven Lab's chief AI scientist for Innovation, Science, and Security and a professor at Stony Brook University, is the principal investigator. Collaborators include Stony Brook University, the New York Power Authority, the Long Island Power Authority, National Grid, and the New York State Energy Research and Development Authority (NYSERDA). More than 150 organizations contribute to the broader GridFM community initiative spanning industry, academia, and government.
"By bringing together the right partners and scaling AI and grid foundation models for the electric grid, we can accelerate planning, improve operations and efficiency, and help build a more affordable, resilient energy system," Hamann said.
Turning a challenge into an advantage
AI is often discussed as a growing burden on the grid due to data center energy demands. Hamann framed the project as a way to reverse that dynamic. "AI is often seen as a growing burden on the electric grid, but through the Genesis Mission we have an opportunity to turn that challenge into an advantage," he said.
Jereza described the practical stakes. "By providing utilities-enhanced AI tools, they can plan and operate the grid faster, more reliably, and more cost-effectively to help meet the demands of a growing economy while delivering affordable, reliable power for homes and businesses," she said.
How the model works
The GridFM is a foundation model - a type of AI trained on broad data that can be adapted to specific tasks. Utilities supply operational data and system experience. Researchers and technology providers contribute AI expertise and computing resources. The model will simulate grid scenarios at a scale that manual planning cannot match, targeting both accuracy and cost efficiency.
Yue Zhao, an associate professor of electrical and computer engineering at Stony Brook, is collaborating with Hamann on the initiative. Stony Brook also recently added another Genesis Mission Phase II project through Yixin "Berry" Wen, an assistant professor developing AI methods for hyperlocal flood-risk information for U.S. energy systems.
Why this matters for science and research professionals
The Genesis Mission is building what DOE calls the world's most powerful integrated science discovery platform, combining AI, supercomputing, quantum systems, and advanced scientific instruments. For researchers in energy systems, computational science, or infrastructure planning, the GridFM project signals where federal funding and institutional priorities are moving: toward large-scale AI for Science & Research applications that pair national labs with university partners. The project's emphasis on operational data from utilities also points to growing demand for researchers who can bridge domain expertise with AI for Research Scientists methods - training foundation models, handling real-world constraints, and validating against physical systems.
Your membership also unlocks:
