Today, the U.S. Department of Energy announced funding for new projects that will harness artificial intelligence to accelerate discovery. Lawrence Berkeley National Laboratory (Berkeley Lab) will lead 13 projects and collaborate as a partner on more than 30 others.
The projects are part of DOE’s Genesis Mission, a historic national initiative that aims to build the world’s most powerful integrated science discovery platform. By uniting government, industry, academia, and philanthropy, it seeks to accelerate breakthroughs in energy, scientific discovery, and national security through a new platform that combines AI, supercomputing, quantum systems, and advanced scientific instruments.
Berkeley Lab’s contributions to the Genesis Mission build on decades of research in high-performance computing, managing large datasets, and pioneering AI models. Researchers will lead new projects across multiple disciplines, advancing fields including critical minerals, materials, manufacturing, fusion, and energy.
“By developing advanced AI tools and combining them with high-quality datasets and high-performance computing, our researchers are maximizing their scientific impact and speeding up discovery,” said Kathy Yelick, director of Berkeley Lab. “Beyond that, the Genesis Mission is also an opportunity to pioneer better AI tools and approaches for all of science, which will bring payoffs for generations to come.”
The goal of the Phase 1 Genesis Mission projects is to identify promising pathways toward transformative scientific capabilities and establish a foundation for future investment and scale. Project teams will design and demonstrate research workflows that integrate AI with scientific investigation, while rigorously evaluating whether those approaches can accelerate discovery, improve predictive capabilities, enhance experimentation, or generate new scientific insights.
The projects that Berkeley Lab leads are:
- AI Decodes DOE Data to Protect U.S. Hydropower and Grid: ASPIRE will create an AI-driven workflow using Department of Energy data to improve precipitation forecasting, helping secure U.S. hydropower generation and protecting critical grid infrastructure.
- AI-Driven Autonomous Labs for Accelerated Materials Discovery: Researchers will create AI systems that work with autonomous laboratories to design, perform, and learn from experiments, dramatically speeding materials discovery and scientific innovation.
- AI-Powered Discovery of Quantum and Optoelectronic Materials: AI powered by physics simulations will model excited states to identify promising quantum and optoelectronic materials, accelerating research for future computing, sensing, and energy technologies.
- AI to Predict Water Supplies for Energy with Novel Precision: Scientists will develop an artificial intelligence framework to jointly predict surface and groundwater dynamics nationwide, supporting energy and water security.
- AI Workflows to Run DOE Codes in the Cloud: The Agentic HPC Pipeline Initiative (AHPI) will build AI workflows that enable U.S. manufacturers to run powerful DOE research codes on commercial cloud platforms.
- Breaking Silos: AI for Nuclear Waste Repository Design: AI will enable a streamlined environment to analyze nuclear waste repository design, dissolving existing silos through a digital-twin approach.
- Closing the Loop: AI-Accelerated Alloy Engineering: Researchers will combine AI structure prediction, high-throughput synthesis, and advanced microscopy to accelerate the optimization and design of complex concentrated alloys for extreme environmental applications.
- HERALD: AI to Unlock Nuclear Archives: The HERALD project will build an AI platform to vet decades of secure government research, unlocking it for the commercial industry to accelerate nuclear innovation.
- Meeting Water-Energy Demand with Local, Actionable Predictions: To predict water resources for energy infrastructure, scientists will use AI to develop a highly accurate prediction system for water availability.
- Novel AI Workflow for Critical Minerals in Geosystems: The team will use AI to optimize recovery of critical minerals from copper, molybdenum, platinum, chrome, and nickel mineral deposits.
- Physics-informed Digital Twins for Fusion Magnet Systems: AI-enabled digital twins will establish a credible path toward the reliable operation of high-temperature superconducting magnets, supporting future compact magnetic confinement fusion power plants.
- Predicting Subsurface Fractures with AI for Energy: Scientists will combine artificial intelligence and geophysical observations to improve prediction of subsurface fractures, advancing geothermal and other energy technologies.
- Safely Storing Nuclear Waste with Advanced 3D Models: Researchers will develop AI-assisted 3D geological models that can effectively identify suitable sites for subsurface nuclear waste storage.
###
Lawrence Berkeley National Laboratory (Berkeley Lab) is committed to groundbreaking research focused on discovery science and solutions for abundant and reliable energy supplies. The lab’s expertise spans materials, chemistry, physics, biology, earth and environmental science, mathematics, and computing. Researchers from around the world rely on the lab’s world-class scientific facilities for their own pioneering research. Founded in 1931 on the belief that the biggest problems are best addressed by teams, Berkeley Lab and its scientists have been recognized with 17 Nobel Prizes. Berkeley Lab is a multiprogram national laboratory managed by the University of California for the U.S. Department of Energy’s Office of Science.
DOE’s Office of Science is the single largest supporter of basic research in the physical sciences in the United States, and is working to address some of the most pressing challenges of our time. For more information, please visit energy.gov/science.
Facts Only
* The U.S. Department of Energy announced funding for new projects harnessing artificial intelligence to accelerate discovery.
* Lawrence Berkeley National Laboratory (Berkeley Lab) will lead 13 projects and collaborate on more than 30 others.
* Projects are part of DOE’s Genesis Mission, an initiative to build a science discovery platform.
* The platform combines AI, supercomputing, quantum systems, and advanced scientific instruments.
* Berkeley Lab's contributions build on research in high-performance computing, managing large datasets, and pioneering AI models.
* Phase 1 of the Genesis Mission seeks to identify pathways for transformative scientific capabilities.
* LEAD projects include: AI Decodes DOE Data to Protect U.S. Hydropower and Grid; AI-Driven Autonomous Labs for Accelerated Materials Discovery; AI-Powered Discovery of Quantum and Optoelectronic Materials; AI to Predict Water Supplies for Energy with Novel Precision; AI Workflows to Run DOE Codes in the Cloud; Breaking Silos: AI for Nuclear Waste Repository Design; AI-Accelerated Alloy Engineering; HERALD: AI to Unlock Nuclear Archives; Meeting Water-Energy Demand with Local, Actionable Predictions; Novel AI Workflow for Critical Minerals in Geosystems; Physics-informed Digital Twins for Fusion Magnet Systems; Predicting Subsurface Fractures with AI for Energy; Safely Storing Nuclear Waste with Advanced 3D Models.
* Berkeley Lab is managed by the University of California for the U.S. Department of Energy’s Office of Science.
Executive Summary
The U.S. Department of Energy has announced funding for new projects utilizing artificial intelligence to accelerate scientific discovery. Lawrence Berkeley National Laboratory (Berkeley Lab) will lead thirteen of these projects and collaborate on over thirty others as part of the DOE’s Genesis Mission. This mission aims to build a comprehensive science discovery platform by integrating AI, supercomputing, quantum systems, and advanced instruments. The goal is to expedite breakthroughs in energy, scientific discovery, and national security through this integrated platform.
Berkeley Lab's work contributes to the Genesis Mission by leveraging existing expertise in high-performance computing, large dataset management, and AI model development across fields like critical minerals, materials, fusion, and energy. The Phase 1 mission focuses on identifying promising research pathways by designing workflows that integrate AI with scientific investigation to accelerate discovery or improve predictive capabilities. Specific projects involve using AI for tasks ranging from improving precipitation forecasting for hydropower security and discovering quantum materials to optimizing the recovery of critical minerals and modeling nuclear waste repositories.
Full Take
The narrative frames AI as a central accelerator for solving massive, interconnected scientific and national security challenges, positioning the Genesis Mission as a necessary platform to synthesize disparate fields (AI, physics, computing) into tangible outcomes in energy and materials science. The structure highlights a methodology: using integrated AI workflows over existing data and computation to achieve predictive accuracy, autonomous experimentation, and knowledge synthesis across complex domains like fusion, critical minerals, and nuclear safety.
The implications suggest a shift from incremental discovery to platform-based acceleration. The specific projects demonstrate an attempt to bridge theoretical physics (fusion magnets) with tangible infrastructure protection (grid forecasting) and resource management (critical minerals). A key tension lies in the balance between speed of discovery and rigorous validation. While the goal is to "accelerate discovery," the success hinges on the quality of the datasets, the fidelity of the AI models, and establishing scientifically sound metrics for evaluating whether an AI-driven workflow genuinely generates *new* insights rather than just optimizing existing processes or introducing algorithmic bias into critical safety calculations.
The pattern observed is a mobilization of massive computational and data resources toward security and energy goals, utilizing cutting-edge techniques. The focus on "breaking silos" via digital twins and unlocking archives suggests an underlying strategic imperative to manage complex, sensitive physical systems. The missing element that requires scrutiny is the mechanism for ensuring the AI tools themselves are vetted for robustness against real-world failure modes, rather than solely focusing on the predictive output generated by those tools.
Bridge Questions: How will the Genesis Mission mandate evaluation protocols to ensure that accelerated discovery does not compromise safety margins in areas like nuclear waste storage or grid stability? What governance structures are being established at Berkeley Lab to manage the integration of diverse AI models across these sensitive domains, and who bears the responsibility for validating those integrated systems? What independent benchmarks will be developed to measure the "transformative scientific capabilities" achieved by these workflows versus traditional research paths?