The Department of Energy (DOE) is aiming to have commercial fusion energy hit the nation’s power grid within the next decade, according to its new Fusion Science and Technology (FS&T) Roadmap.
Unveiled Thursday, the FS&T Roadmap is an effort by the DOE to carry out an executive order signed by President Donald Trump on his first day in office to “unleash” American energy.
The roadmap is built on three pillars: building critical infrastructure; innovating through advanced research, high-performance computing, and artificial intelligence; and developing public-private partnerships, regional manufacturing hubs, and growing the workforce.
“The Fusion Science and Technology Roadmap brings unprecedented coordination across America’s fusion enterprise,” said Darío Gil, undersecretary for science at DOE. “For the first time, DOE, industry, and our National Labs will be aligned with a shared purpose – to accelerate the path to commercial fusion power and strengthen America’s leadership in energy innovation.”
The roadmap focuses on “AI-Fusion convergence,” which the DOE said is “a national capability that will weave through all DOE fusion program elements.”
According to the roadmap, the American Science Cloud (AmSC) – a multimillion-dollar investment to structure and process data for AI – will use AI-driven data sharing, advanced computing, and foundation model development. AmSC’s Fusion Energy Sciences teams are building models for materials, digital twins of DOE and private fusion facilities, and AI tools to enhance high-performance computing software.
Meanwhile, in partnership with NVIDIA and IBM, the Princeton Plasma Physics National Laboratory is developing Stellar-AI, a next-generation supercomputing cluster optimized for fusion. This cluster will serve as a collaborative hub for academia, industry, and national labs to train foundation models, according to the roadmap.
“This era is characterized by strong alignment between the public sector roadmap and the private sector’s stated ambitions to deliver fusion power on an aggressive timeline and is increasingly enabled and accelerated by the revolutionary potential of AI-Fusion convergence,” the DOE said.
The department did note that while it intends to carry out the goals and meet the timelines established in its roadmap, work “is contingent on the development of future public– private partnerships.”