DOE Awards $68 Million for AI-Driven Scientific Research Across 43 Projects
March 14, 2026 · 2 min read
Jared Klein
The Department of Energy is betting $68 million that artificial intelligence can accelerate scientific discovery — and it's spreading that bet across 43 awards covering everything from privacy-preserving machine learning to energy-efficient computing hardware.
The funding announcement covers 11 multi-institution projects selected through competitive peer review under DOE's Funding Opportunity Announcement for Advancements in Artificial Intelligence for Science. Projects run up to three years, with $20 million allocated from FY 2024 dollars and remaining funding contingent on congressional appropriations.
Where the Money Goes
The projects cluster around three technical frontiers. First, foundation models: teams will study how large scientific AI models improve as they scale in size and complexity, mirroring the capabilities explosion seen in commercial AI but directed at scientific applications. Second, distributed learning: researchers will develop techniques to train models across multiple institutions while preserving data privacy — a critical requirement for sensitive research data in energy, health, and national security. Third, efficiency: projects will develop energy-efficient algorithms designed to run on next-generation microelectronics, addressing the growing concern that AI's energy appetite could undermine its scientific benefits.
The Policy Backdrop
The funding responds to Executive Order 14110 on safe, secure, and trustworthy AI development. It arrives as DOE's Office of Science received a nearly 2 percent budget increase to $8.4 billion in FY 2026 — one of the few federal science agencies to see growth rather than cuts. The Advanced Research Projects Agency-Energy (ARPA-E) was not as fortunate, receiving $350 million after Congress rejected the administration's proposed 57 percent reduction but still imposed a 24 percent cut from FY 2025 levels.
What AI Researchers Should Know
The 43 awards span multiple national laboratories and universities, creating collaboration opportunities for researchers who weren't on the initial project teams. DOE frequently issues supplemental calls and lab-directed research partnerships that build on funded projects. Scientists working at the intersection of AI and physical sciences should monitor the Advanced Scientific Computing Research program page for follow-on opportunities. The in-depth analysis on the Granted blog covers the full scope of federal AI funding programs now open for applications.