The U.S. Department of Energy (DOE) has launched the Genesis Mission, a national initiative designed to secure the United States’ supply of critical minerals by integrating artificial intelligence (AI), supercomputing, quantum systems and advanced scientific instruments into a discovery platform.
The initiative aims to increase mineral independence and national security by accelerating the discovery of domestic deposits and improving the recovery of materials from waste. According to the DOE, the first phase of funding is intended to establish a foundation for future investment and scale while identifying promising research pathways.
AI Framework for Mineral Exploration
As part of the Genesis Mission, the DOE has awarded funding for a project known as GEM-AI. Led by Artur Dubrawski, the Alumni Research Professor of Computer Science in the Robotics Institute at Carnegie Mellon University’s (CMU) School of Computer Science, the project will develop an agentic AI system to identify critical minerals in the ground.
The GEM-AI framework will integrate diverse scientific data, including:
- Metagenomic analysis and geologic sampling data
- Sensor imagery and aerial photography
- Existing satellite photography
The system will search for geologic and biological markers, as previous research indicates that materials near the surface or floating in dust particles can signal valuable deposits below. Dubrawski stated that the goal is to reduce the cost, time and resources required for exploration within the U.S.
The research team includes experts from the Carnegie Mellon University Auton Lab, the Colorado School of Mines, and the Pittsburgh Supercomputing Center.
Recovering Metals from Battery Waste
This project, titled A Multi-agent AI Framework for Discovering Chemical Drivers of Selective Critical Metal Recovery from Complex Battery Waste,
is a collaboration with the University of Southern California and the SLAC-Stanford Battery Center.
The project targets metals such as manganese, nickel and cobalt, which are essential for battery manufacturing. Because a single spent electric vehicle battery can contain tens of pounds of these metals, recovering them domestically could reduce reliance on imports.
To overcome the current challenges of high chemical use and extensive trial and error, the team is building a multi-agent AI system. These specialized agents will draw on knowledge from geology and biochemistry to propose recovery strategies and evaluate new chemical pathways.
Ahamed Irshad Maniyanganam, the lead researcher and SLAC associate scientist, stated that the team will run several AI-experiment cycles over nine months. The primary objective is to recover target metals at 80 percent purity or better, benchmarking these results against conventional recovery methods and literature searches.
Broad Scientific Scope and Infrastructure
The Genesis Mission represents a broader effort to create an integrated science discovery platform. In addition to the battery recycling project, the SLAC National Accelerator Laboratory is participating in 10 other Phase I projects. These efforts span several scientific domains, including:
Project teams across the mission are tasked with designing research workflows that integrate AI with scientific investigation. According to the DOE, these teams must rigorously evaluate if their approaches improve predictive capabilities, enhance experimentation, generate new scientific insights or accelerate the overall pace of discovery.
Co-investigators on the battery recovery project include Jayakanth Ravichandran of the USC Viterbi School of Engineering, Jagjit Nanda of the SLAC-Stanford Battery Center, and Frank Abild-Pedersen of the SLAC-Stanford SUNCAT Center for Interface Science and Catalysis.
