Brookhaven National Laboratory and Amazon Web Services partner to connect AI data center infrastructure.

The U.S. Department of Energy's Brookhaven National Laboratory and Amazon Web Services have announced a partnership to accelerate GridSearch, an AI-driven solution designed to help new generation facilities connect to America's electric grid. The partnership was unveiled during the AWS Summit in Washington, D.C.

GridSearch leverages AWS's cloud computing infrastructure and AI capabilities to optimize the grid connection process. The project employs a Grid Foundation Model (GridFM), developed as an open-source initiative under the Linux Foundation Energy with major contributions from IBM Research, Hydro Quebec, Brookhaven National Laboratory, Stony Brook University, Argonne National Laboratory, and others. GridFM enables significantly faster grid simulations while maintaining accuracy and robustness.

"GridSearch is an excellent example for AI enabling AI — what we call 'AI4AI,'" said Hendrik Hamann, chief AI scientist for Innovation, Science and Security at Brookhaven National Laboratory and professor in the School of Marine and Atmospheric Sciences at Stony Brook University. "It aims to provide critical decision support to stakeholders and help them accelerate the data center interconnection process with optimal affordability and minimal disruption to the electric grid."

Brookhaven Lab Director John Hill framed the partnership within DOE's Genesis Mission, a national initiative to build a leading scientific platform for discovery science, national security, and energy innovation. "GridSearch demonstrates how we can apply Genesis Mission capabilities to real-world applications, helping the nation achieve AI leadership and energy dominance," Hill said.

Stony Brook University has contributed significantly to the Genesis Mission through 89 proposal submissions, including 28 joint collaborations with Brookhaven National Laboratory.