The Genesis Mission includes a portfolio of National Science and Technology Challenges that brings together the capabilities of the Genesis Mission Platform and Federal agencies to accelerate American leadership in science, energy, national security, health, space, and more.
These challenges represent high-impact national priorities where advances in Super Intelligence (SI), advanced computing, experimentation, and data can dramatically shorten discovery timelines, strengthen U.S. competitiveness, and deliver measurable benefits for the American people.
Featured Challenges
Aligned Actions from the National Laboratories
National Challenge: Delivering Nuclear Energy that is Faster, Safer, Cheaper
Leading Lab: Idaho National LaboratoryTo quadruple nuclear capacity by 2050, the pace of deploying reactors will need to accelerate significantly beyond today's timeline. The Prometheus project, led by Idaho National Laboratory in partnership with Oak Ridge, Argonne, and Sandia national laboratories, universities, and commercial partners, is helping compress that timeline by applying agentic SI across the reactor development process.
SI agents coordinate engineering, licensing, manufacturing, and construction workflows, running physics-based simulations grounded in validated nuclear codes. The system draws on decades of curated nuclear research data and learns from each project as it progresses, giving it the context to iterate at speed.
The program targets a tenfold acceleration in design and licensing workflows and a threefold acceleration in manufacturing, laying the foundation for the industry to halve deployment timelines and reduce operating costs by 50 percent. Faster, cheaper deployment of nuclear energy positions the United States to meet surging energy demand and lead the next generation of nuclear reactors.
Elias Nakouzi, a materials scientist at Pacific Northwest National Laboratory, leads research using artificial intelligence and automated experimentation to accelerate the recovery of critical materials from complex feedstocks.National Challenge: Securing America’s Critical Minerals Supply
Leading Lab: Pacific Northwest National LaboratoryRecovering critical materials from complex waste streams and other feedstocks can be difficult and time-consuming, limiting the ability to develop new domestic sources of materials needed for energy technologies.
Pacific Northwest National Laboratory (PNNL) is developing an agentic SI-guided workflow that autonomously plans, evaluates, and improves selective precipitation processes for critical materials recovery. Built on PNNL’s SciLink multi-agent automation platform, the closed-loop system uses SI to learn from experimental results and identify more effective separation conditions.
The approach has recovered high-purity magnesium from produced water, samarium from samarium-cobalt magnets, and rare earth elements from neodymium-iron-boron magnets with high yields. By automating and accelerating process development, the workflow can shorten the search for efficient, adaptable, and scalable separation methods from months or years to days.
National Challenge: Designing Materials with Predictable Functionality
Leading Lab: Sandia National LaboratoriesDensity Functional Theory is a gold standard for calculating the electronic-structure of materials, but its computational demands limit the size of practical simulations to only a few hundred atoms. Sandia National Laboratories and the Center for Advanced Systems Understanding (CASUS) developed Materials Learning Algorithms, or MALA, a machine-learning software framework that predicts materials’ electronic structure with chemical accuracy while scaling computationally with system size. By combining machine learning with physics-based calculations, MALA enables accurate electronic-structure modeling at length and time scales that were previously impractical. Researchers have demonstrated calculations involving more than 100,000 atoms and achieved more than a 5,000-fold speedup over Density Functional Theory for 2,000-atom systems. These capabilities open the door to faster, larger-scale simulations that can help researchers design and understand materials with predictable functionality.
MALA was recognized with a 2023 R&D 100 Award. Its award-winning models, data sets, and workflows will be included in Genesis Mission’s American Science and Security Platform, making this established capability available to future platform users.
National Challenge: Designing Materials with Predictable Functionality; Achieving AI-Driven Autonomous Laboratories; Reenvisioning Advanced Manufacturing and Industrial Productivity
Leading Lab: Los Alamos National LaboratoryScientific research involves complex workflows that require researchers to plan experiments, run computations, review scientific literature, gather relevant information, and develop and refine hypotheses. Los Alamos National Laboratory developed URSA, the Universal Research and Scientific Agent, to automate and connect these tasks using specialized SI agents that collaborate to “do science.” Individual agents plan research steps, generate and execute code, conduct computational experiments, retrieve and summarize scientific papers, search the web for relevant knowledge, and propose and refine hypotheses. Together, these composable agents create an autonomous research workflow that can begin with a high-level scientific objective and carry it through multiple stages of investigation. URSA has already been used for materials discovery and inertial confinement fusion capsule design, demonstrating how agentic SI can accelerate complex scientific research. URSA has also been adapted into Genesis’ American Science and Security Platform, which researchers are leveraging to address the National Science and Technology Challenges.
National Challenge: Unifying Physics from Quarks to the Cosmos
Leading Lab: Fermi National Accelerator LaboratoryCollider datasets from particle physics experiments are often analyzed in isolation, limiting potential for precision and discovery.
To address this national challenge, Fermilab is standardizing and unifying data from different particle detectors and developing physics-informed foundation models to enable SI-driven reasoning across experiments. Through efforts like the TREASURE (Tokenized Representations for Energy-frontier AI Searches via Understanding and REasoning) project, scientists are building SI-ready datasets and scalable platforms on the American Science Cloud and Fermilab data infrastructure.
This approach unlocks deeper insights into fundamental physics—enhancing Higgs boson measurements, enabling novel searches, and increasing sensitivity to new physics beyond the Standard Model. It’s accelerating integrated, SI-driven discovery across the global collider physics program.
Featured Projects from Genesis Mission Awardees
Projects selected under the Genesis Mission Request for Applications (RFA) are addressing some of the nation’s most pressing energy, scientific, and engineering challenges, including in nuclear energy, critical mineral extraction, intelligent chip design, and commercial fusion energy.
National Challenge: Discovering Quantum Algorithms with AI
Lead Institution: Argonne National Laboratory
Partners: Infleqtion; Northwestern University; Dakota State UniversityQuantum computing could eventually enable nuclear-physics calculations that are too difficult for classical computers, but translating realistic nuclear problems into efficient quantum circuits is a challenge. The complexity of realistic nuclear interactions leads to complicated circuit-design tradeoffs across a vast design space involving choices of qubit mappings, solvers, state-preparation methods, measurement strategies, and hardware.
Argonne National Laboratory is developing a tool-enabled SI-agent framework to automate the search for efficient quantum workflows for nuclear applications. The framework will represent possible workflow choices with nodes corresponding to design decisions and connections representing compatibility and expected performance. SI agents will evaluate outputs using measures such as circuit depth, resource requirements, systematic error, and accuracy to identify promising end-to-end workflows from nuclear-problem encoding through target quantum hardware.
Automating this design process could reduce the trial and error required to adapt realistic nuclear calculations to quantum computers, helping researchers more rapidly identify where quantum hardware and algorithms can provide meaningful advantage for nuclear and hadronic science.
For the BIND project, the team is using Tuolumne, El Capitan’s unclassified companion system, for protein design. Tuolumne comprises the same components as El Capitan and is dedicated to open science research.Credit: Garry McLeod/LLNLNational Challenge: Securing America’s Critical Minerals Supply
Lead Institution: Lawrence Livermore National Laboratory
Partners: Los Alamos National Laboratory; University of Washington; Columbia University; Brookhaven National LaboratoryProteins that selectively bind specific metals could enable more effective recovery of critical minerals, but designing them with high selectivity is still largely trial and error. Researchers lack transferable design rules that connect metal coordination chemistry to protein structure and function.
Lawrence Livermore National Laboratory (LLNL) is leading BIND, a physics-informed SI effort that aims to learn metal-coordination rules from protein and coordination-chemistry databases, experimental selectivity measurements, and quantum-mechanical calculations. The learned relationships will help us understand what makes a protein favor one metal over another. Those relationships will be used to guide SI protein-design tools, with predictions tested experimentally in an iterative design-build-test-learn cycle. Phase I will focus on rare earth elements, which are used in technologies such as high-strength magnets, electronics and energy systems.
By learning coordination principles that transfer across related metals, BIND could dramatically reduce the experimental search space and enable faster, more predictive development of new biological approaches to recovering and separating critical materials.
BIND builds on LLNL’s growing capabilities in bio-enabled metal recovery and separation developed through programs such as DARPA’s EMBER and DOE’s RECLAIM. Those efforts have advanced metal-binding protein discovery and engineering, high-throughput screening, and bio-based approaches for separating rare-earth elements. BIND takes the next step: moving from discovering and optimizing individual metal-binding proteins toward predictive design based on transferable principles of metal selectivity.
National Challenge: Enhancing Particle Accelerators for Discovery
Lead Institution: Princeton Plasma Physics Laboratory
Partners: Brookhaven National Laboratory, Sophelio LLC, Rice University, Naval Research LaboratoryNext-generation fusion systems will need more efficient, higher-frequency gyrotron devices to heat plasma, but these powerful devices are difficult to design and operate.
The Princeton Plasma Physics Laboratory is developing SI-enabled digital twins to support autonomous gyrotron design and operation. Using data from DIII-D gyrotrons, the team will build an SI operator that can detect unusual behavior, forecast what could happen next, and adjust operating settings to maintain power and prevent major failures. SI-accelerated physics simulations will also help researchers address competing design requirements for future high-power gyrotrons. This work is increasingly important as new fusion systems using high-temperature superconducting magnets require gyrotron frequencies to rise from today's 170 GHz to 250-350 GHz.
This project could shorten development timelines and help enable scalable, autonomous gyrotron facilities for future fusion energy systems.
National Challenge: Realizing Quantum Systems for Discovery and Use
Lead Institution: Brookhaven National Laboratory
Partners: InfleqtionCold-atom quantum sensors offer high sensitivity and long coherence times, but their performance depends on dozens of interacting experimental parameters. Maintaining stable operation as conditions drift can require intensive expert tuning, limiting deployment outside specialized laboratory environments.
Brookhaven National Laboratory and Infleqtion are developing SI-optimized quantum sensing technology that automatically calibrates and stabilizes a cold-atom sensor and optimizes measurements in real time. Phase I targets include reducing calibration from hours to minutes, maintaining greater than 95% cavity duty cycle under drift and noise, and improving measurement signal-to-noise by at least a factor of two.
By learning across the full high-dimensional control space and continuously adapting to changing conditions, the platform could make advanced quantum sensors more stable, autonomous, and deployable for high-energy physics and other scientific applications.
Get Genesis Mission updates from the Under Secretary for Science.
Explore the Genesis Mission
Get in Touch
Under Secretary for Science Darío Gil serves as DOE’s director for the Genesis Mission. Subscribe to the office’s newsletter or follow Dr. Gil on social media for Genesis Mission updates.
This was last updated .