5 projects at UW–Madison aimed at transforming science and energy with AI receive DOE Genesis Mission funding

The projects, spanning fusion to critical minerals, are among 278 funded awards to national labs, companies, universities and non-profits in order to accelerate breakthroughs in energy, scientific discovery and national security through a new platform that combines AI, supercomputing, quantum systems and advanced scientific instruments.

Researchers from the University of Wisconsin-Madison joined industry partners and colleagues from universities and national laboratories at the first annual Genesis Mission Summit in Washington, D.C., on Wed. July 22 as the U.S. Department of Energy announced the first phase of projects funded through the Genesis Mission.

UW-Madison will receive five Phase I awards, ranging from $500,000 to $750,000, to support nine-month projects that are aimed at addressing some of the nation's most pressing energy, scientific, and engineering challenges, including in nuclear energy, critical mineral extraction and commercial fusion energy.

UW-Madison has earned more awards than any other public university in the nation, tying with Stanford University and coming in just behind the six granted to the Massachusetts Institute of Technology in terms of overall number of awards.

The agency granted 278 awards to 168 universities, 87 national labs, 19 companies and four non-profits. The current awards emerged from a funding opportunity DOE announced in March 2026, which the agency says generated the largest response to a funding opportunity in DOE history.

The Genesis Mission brings together DOE's national laboratories, universities, industry and other partners to collaborate on efforts spanning energy, science and national security. It is focused on using AI and advanced computing to support scientific research and deliver tangible benefits to society.

"The Genesis Mission recognizes that today's most complex scientific and technological challenges cannot be solved by any one institution alone. Each partner brings unique strengths that, together, can accelerate discovery and innovation," says Dorota Brzezinska, UW-Madison's vice chancellor for research.

"We are proud to have UW-Madison recognized at the genesis of Genesis for its research strengths and the opportunity to contribute novel partnerships and workforce development approaches for the benefit of the nation," Brzezinska adds. We are translating research into real-world impact, true to the Wisconsin Idea, and preparing the next generation of scientists and engineers with the skills needed to lead and succeed in a future with AI."

Funding to support a variety of innovative projects at UW-Madison

Genesis Mission Phase I awards support teams as they design and test research workflows that integrate AI with scientific investigation. The projects selected for funding span the College of Engineering, the College of Letters & Science and the College of Computing & Artificial Intelligence. They include:

Building foundation models for physics

Led by Mariel Pettee, professor of physics in L&S, in collaboration with researchers from Argonne National Laboratory, Carnegie Mellon University and the University of Kentucky, the project will investigate how large AI models can learn from and connect data across multiple areas of physics.

In particular, the project will explore how machine learning, a subset of AI in which computers "learn" from data, can help reveal new, previously overlooked connections.

"As a new faculty member, it means a lot to have support for curiosity-driven fundamental research," says Pettee, who joined the UW-Madison physics faculty as part of the university's RISE-AI initiative in August 2025. "This project is remarkable, because it brings together disciplines and institutions that would not typically work closely together."

Accelerating critical mineral recovery with AI

Athena Nghiem, professor of geoscience in L&S, will lead a project to develop an AI system designed to dramatically speed up the recovery of copper and other critical minerals from U.S. deposits that are currently difficult or uneconomical to access.

Her collaborators include industry partners Ekion Pty Ltd. located in Australia and Thunderstone Inc. in San Francisco, both specialized in extracting critical metals from rock.

"We will develop an agentic AI framework, a team of specialized AI agents that work together like an autonomous research group," Nghiem says. "The system will analyze scientific literature, run complex simulations, compare results with laboratory and field data and continuously refine its models - compressing months of work into days. I am especially excited for this incredible opportunity to build this important research direction with industry collaborators."

Advancing fusion energy through real-time AI

David Smith, a scientist in nuclear engineering and engineering physics in the College of Engineering, will lead a project in collaboration with Semin Joung, UW-Madison nuclear engineering and engineering physics scientist, and Ryan Coffee, a scientist at SLAC National Accelerator Laboratory.

Their work is focused on advancing fusion energy by using advanced AI systems to process large streams of experimental data in real time at the DIII-D National Fusion Facility in San Diego, California. Their approach is intended to enhance plasma control, a critical part of magnetic fusion energy.

"This funding allows our team to rapidly validate the technologies and predictive models needed to bring commercial fusion energy closer to reality," Smith says.

Speeding fusion reactor design

Paul Wilson, professor of nuclear engineering and engineering physics in the College of Engineering, will lead a project aimed at improving fusion energy systems by advancing a critical component of a fusion reactor called a breeding blanket. The work will rely on both AI and machine learning to accelerate improved breeding blanket design.

The goal is to reduce the time it takes to design an optimal configuration.

"With that improved performance methodology, we will determine optimal blanket configurations … developing a large data set used to train machine learning models that can further accelerate the optimization process," says Wilson.

Improving AI performance on supercomputers

Matt Sinclair and Shivaram Venkataraman, computer sciences professors in the College of Computing & Artificial Intelligence, are working together to design more efficient supercomputers, which is particularly important given the energy demands of AI and large-scale computing systems.

"We sit at this really interesting intersection of how we design the hardware to run these (high-powered computing) systems efficiently, how we build these systems to run them effectively, and how the domain scientist doing physics, chemistry, and meteorology, for example, try to get their workloads to run effectively on these systems," says Sinclair.

Their project will take place in collaboration with a variety of subject matter experts from the University of Virginia, Fermilab and Oak Ridge National Lab.

"It's really exciting because it gives us the opportunity to both come up with new ideas, but also to execute them and build concrete prototypes, systems and products," says Venkataraman. "This is the beginning - Phase 1 - and our hope is that we recruit students and other scientists to work with us on these problems and hopefully our deliverables will help us put together a strong application for Phase 2. It's exciting for students here - a unique opportunity to be able to work on projects like this."

Subaward to UW-Madison in its own backyard

A number of UW-Madison researchers are also participating as sub-awardees in several other Genesis Mission projects. One example is a fusion energy project led by UW-Madison spinoff company Realta Fusion, in collaboration with Cary Forest, a professor of physics in L&S. The effort builds on a longstanding collaboration between Realta Fusion and researchers at UW-Madison.

According to Devesh Ranjan, Grainger Dean of the College of Engineering, working alongside industry partners like Realta yields significant benefits "because those collaborations help accelerate the path from breakthrough discoveries to real impact."

And university investments help make opportunities like this possible. For instance, he says, "through our Engineering Moonshots initiative, we have invested in areas of growing national importance - including fusion energy, AI-enabled autonomous systems and critical minerals - and these awards reinforce the strength of that strategy."

Founding Dean Remzi Arpaci-Dusseau in the College of Computing & Artificial Intelligence emphasizes that the awards reinforce the important role universities play in tackling some of society's greatest challenges.

"Artificial intelligence is one of the most powerful tools we have for advancing scientific discovery, but its greatest impact comes when it's paired with deep expertise across disciplines," Arpaci-Dusseau says. "These awards highlight the strength of UW-Madison researchers working together to solve complex challenges, and they reflect the mission of the College of Computing & Artificial Intelligence: advancing people and technology that expand what's possible for society."

For interim L&S Dean Kristin Eschenfelder, these awards confirm that the university is living up to its reputation of rigorous research to benefit the public good.

"Pushing scientific boundaries is our mission at UW-Madison and the College of Letters & Science," Eschenfelder says. "Using AI to advance science faster in fields like physics and geoscience is a core goal of the RISE-AI initiative, and a central priority of the university, where we support curiosity driven discovery to grapple with today's complex challenges. This is great evidence of success."