Artificial Intelligence
Artificial Intelligence (AI) is transforming everyday life — and the way that science is carried out. Brookhaven Lab is at the forefront of this revolution. Researchers apply AI across the Lab’s wide range of scientific initiatives to accelerate discoveries, and they’re developing new technologies to improve AI itself.
Brookhaven Lab is advancing the goals of the Genesis MissionGenesis Mission, a flagship initiative led by the U.S. Department of Energy to dramatically accelerate American scientific discovery, strengthen U.S. economic leadership, and enhance national security through AI. Researchers are developing new ways to improve AI for science and society, and they’re applying AI across the Lab to accelerate discoveries.
The National Science and Technology Challenges are 26 high-impact technical problems aligned to urgent national priorities. Multidisciplinary teams at Brookhaven are applying their expertise in AI and leveraging the Lab’s world-class capabilities to tackle many of the National Challenges.
The American Science and Security Platform connects the nation’s best computing resources, scientific data, and research facilities to unify access to AI. Brookhaven Lab supports the platform by participating in the American Science Cloud and the Transformational AI Modeling Consortium.
Enhancing Scientific Imaging
Researchers at Brookhaven Lab are using AI to conduct imaging studies like never before. Complex materials like microelectronics and sensitive biological samples like proteins are notoriously difficult to capture at the ultra-small scales where new discoveries emerge. By training and integrating AI models at the Lab’s world-leading imaging facilities, scientists can extract insights from incomplete data, detect subtle patterns, and shorten analysis times from weeks to minutes.1
Discovering New Medicines
Brookhaven Lab operates AI-enhanced facilities to accelerate the search for new medicines. Finding treatments for cancer and other diseases currently relies on time-intensive experiments, and AI models lack sufficient data to be a reliable improvement. The Lab has equipped its cutting-edge biotechnology facilities with AI and robotics to develop a continuously operating discovery platform that generates large, high-quality datasets for training the next generation of AI models.2
Expanding the Electric Grid
As the demand for electricity grows, especially from facilities like data centers, researchers at Brookhaven Lab are developing AI models to expand the power grid rapidly and efficiently. Traditional methods of simulating the electric grid are too slow for today’s needs. Brookhaven’s AI approach is more than 100 times faster, eliminating the need for large infrastructure investments and improving the grid’s resiliency and affordability.3
Discovering New Materials
Scientists at Brookhaven Lab are using AI to discover new materials for applications in energy, quantum technologies, and advanced electronics. Instead of using multiple techniques to test thousands of material variations one at a time, the Lab’s researchers use AI to coordinate simultaneous experiments that autonomously guide one another. AI models at the Lab also analyze data in real time, focusing only on the most informative measurements. These methods reduce experiment runtimes from months to days — with over 99% accuracy.4
Supplying Critical Minerals
Brookhaven Lab’s AI research is strengthening the domestic supply chain of critical minerals. Many technologies, including electronics and energy systems, are made of minerals that are costly and complex to secure, so the U.S. currently relies on foreign supply chains. The Lab’s scientists are developing novel AI models that can design biomolecules to efficiently extract, separate, and concentrate these materials, simplifying the process from more than 500 required steps to less than five. The Lab is also developing an AI-driven autonomous laboratory to demonstrate this approach in practice.5
1 Supports the “Enhancing Particle Accelerators for Discovery” and “Recentering Microelectronics in America” National Challenges.
2 Supports the “Enhancing Particle Accelerators for Discovery” and “Scaling the Biotechnology Revolution” National Challenges.
3 Supports the “Securing U.S. Leadership in Data Centers” and “Scaling the Grid to Power the American Economy” National Challenges.
4 Supports the “Enhancing Particle Accelerators for Discovery” and “Designing Materials with Predictable Functionality” National Challenges.
5 Supports the “Securing America’s Critical Minerals Supply” National Challenge.
Brookhaven's AI Strategy
Inventing the future of scientific discovery with AI
Brookhaven is defining the next generation of scientific research by pioneering an ecosystem where AI, advanced research tools, and human ingenuity work together. The goal is to transform vast scientific data into knowledge, knowledge into understanding, and understanding into solutions for society's greatest challenges. Brookhaven Lab’s AI strategy reimagines how science is done and aspires to expand what humanity can discover.
The Exocortex
Researchers at Brookhaven Lab are building a “science exocortex,” an ecosystem of AI agents that collaborate to accomplish complex tasks. Individual AI agents are trained to carry out specialized tasks, such as searching scientific literature, generating a hypothesis, designing experiments, operating research instruments, and analyzing data — all simultaneously, without human intervention. The exocortex, named for its design goal of simulating a new extension of the human brain, will deliver new insights to a human researcher through natural conversation.
Embodied AI
Robots are regularly used in science for simple tasks like moving a sample from one known location to another. Brookhaven Lab is at the frontier of applying robotics to more complex tasks, providing AI agents with a sophisticated, dexterous “body” to carry out experiments and lab operations entirely on their own. The Lab takes an AI-first approach to this field of “embodied AI,” focusing on improving the algorithms that will make embodied AI successful.
Autonomous Experiments
Brookhaven is a world-leader in developing and using AI to autonomously run scientific experiments — helping researchers by conducting experiments, analyzing the results in real time, and deciding which experiments to run next. Such agents are already in routine use at the Lab’s National Synchrotron Light Source II, dramatically increasing experimental efficiency and freeing up researchers’ time for higher-level thinking and making discoveries.
Particle Accelerator Operations
The Lab is home to multiple particle accelerators that generate massive datasets. Researchers use all this data to train generative AI models that can, in turn, enhance the operational efficiency and reliability of these accelerators. AI surpasses humans in fault identification and optimizing accelerator performance in real time. Researchers will also use AI to enhance their search for rare particle interactions at the Lab’s future Electron-Ion Collider.
AI in Hardware
Brookhaven Lab is developing custom computer chips with built-in AI, enabling data to be processed right where it’s collected instead of sending it elsewhere. These chips use advanced components designed to mimic how the human brain works, along with light-based and traditional electronic technologies, for energy-efficient applications, including next-generation accelerators.
Digital Twins
Researchers can make digital replicas of real-world systems, called digital twins, that use AI to simulate responses and behaviors under more secure, safe, and cost-efficient conditions. Brookhaven Lab is researching efficient supercomputer methods to quickly render accurate digital twins and quantify inaccuracies.
Training AI for Science
Scientific data is more complex than the text and images that most AI models have been trained on. Brookhaven Lab is exploring how researchers might apply large language model techniques to scientific data for the next generation of AI models. Such models would be transformative for the scientific enterprise and would have applications outside of scientific research — for example in medicine, manufacturing, and agriculture.
Safe, Secure, Trustworthy AI
For scientific and national security uses, it is essential that AI is safe, secure, and trustworthy. Brookhaven has developed privacy-preserving AI (PPAI) technologies and AI inferences that are fully encrypted. Brookhaven has demonstrated these and other methods in areas including Arms Controls and Verification and Nuclear Safeguards. Such PPAI technologies can also benefit non-security applications where privacy is important, including in healthcare and the financial industry.
Embodied AI at Brookhaven Lab
December 15, 2025
Researchers at Brookhaven Lab are making progress in many areas of artificial intelligence (AI), including “embodied AI” — systems that incorporate AI into a physical object, such as a robot, that can interact with and learn from the physical world. Carlos Soto, from the Lab's Computing and Data Sciences Directorate, leads Brookhaven's work in embodied AI. Hear him explain this fascinating research and where it's leading.
Latest AI research news
Brookhaven Lab Intern Uses Artificial Intelligence to Train Robots
Jasmin Lin explored artificial intelligence and robotics through hands-on internship experiences at Brookhaven National Laboratory.
AI Shapes the Design of the Electron-Ion Collider
Artificial intelligence and machine learning are shaping major design and research decisions for the planned Electron-Ion Collider, a next-generation nuclear physics research facility that will collide electrons with protons or nuclei to probe matter’s structure.
Energy Undersecretary Dario Gil talks Genesis Mission, Brookhaven's Grid FM AI work
Dario Gil lays out vision to double the productivity of U.S. research and development within a decade.
Machine Learning Speeds Up Search for Better Catalysts
A new framework approach offers reliable catalyst screening and sheds light on key chemical reaction steps.
At Brookhaven's National Synchrotron Light Source II, researcher Andi Barbour uses AI to improve the quality of her data. Barbour’s team uses X-rays to study how materials change over time, but detector noise and beam fluctuations make traditional analysis slow and unreliable. Their AI agent automatically cleans data and extracts key patterns that reveal structural changes. This automation greatly reduces computation time, increases accuracy, and provides clearer, high-resolution insights that can be directly compared with models.
