Today, U.S. Department of Energy (DOE) labs house four of the ten fastest and most powerful supercomputers in the world, uniquely positioning DOE to push the limits of artificial intelligence (AI).
DOE’s Artificial Intelligence & Technology Office (AITO) was established to harness the Department’s world-class leadership in high-performance computing, facilities, and team science to build upon and accelerate America’s existing AI research in all of these areas and more. Today, AI is being applied across the Department to accelerate the pace of discovery in a wide variety of areas such as:
Energy
- Idaho National Lab (INL) researchers are also using AI to discover better ways to turn organic matter into energy.
- Multiple DOE Labs are using AI technologies and tools to prevent, detect, and effectively respond to electric grid disruptions, strengthening the resiliency of our grid.
- National Energy Technology Lab (NETL) is expanding its work to develop the next generation of methane leak detection technology using an AI aerial platform.
- DOE is also supporting an effort to achieve higher efficiency coal plants using AI techniques.
National Security and Emergency Response
- INL has developed an autonomic intelligent cyber sensor, which gives industries the power to quickly identify and divert hackers, using machine learning to identify and map industrial control systems.
- SLAC National Accelerator Laboratory and Berkeley Lab are using AI to prevent or minimize electric grid failures and bounce back faster from storms, solar eclipses, cyberattacks and other disruptions. Los Alamos Lab is analyzing massive amounts of seismic data to help them better understand earthquakes, anticipate how they will behave, and provide quicker and more accurate early warnings.
- Scientists are using AI to better understand research that could help assure the safety, security and effectiveness of the U.S. nuclear deterrent. Researchers at Pacific Northwest National Lab (PNNL) are using AI to strengthen nuclear non-proliferation by testing signals of potential significance.
Materials Science
- Ames Lab’s CaloriCool consortium is using AI to screen new materials that could be used to radically improve the energy efficiency of refrigeration technologies.
- Scientists at Fermilab are using machine learning (ML) to search for new particles and develop a deeper understanding of fundamental forces, such as neutrinos.
- SLAC researchers are using AI and accelerated experiments to speed the discovery of metallic glasses, materials that are stronger and more efficient than today’s best steel, at a fraction of the time and cost required for conventional discoveries.
Health Care
- Argonne, Los Alamos, Lawrence Livermore, and Oak Ridge Lab scientists are using AI and ML to develop new approaches to predicting and treating cancers as part of the CANDLE (CANcer Distributed Learning Environment) consortium.
- Researchers are using AI to create more accurate representations of pathology reports and interpretations of mammograms, improve vaccination campaigns, diagnoses, treatments and outcomes for Americans, from children to veterans.
- Scientists at Lawrence Livermore, Berkeley, and Argonne Labs, as well as DOD are using AI and DOE’s supercomputing capabilities to improve understanding of, and develop better treatments for, traumatic brain injuries.
Transportation
- Researchers at Argonne, Oak Ridge, and Sandia Labs are using AI and ML to address complex transportation problems, including reducing traffic jams, improving fuel efficiency and predicting how transportation will evolve in the future.
- Researchers at Berkeley Lab have launched projects to apply AI to self-driving cars in order to smooth traffic, reduce fuel consumption and improve air quality predictions.