AI Partnership Aims to Halve Nuclear Reactor Costs and Build Times

The Idaho National Laboratory (INL) and NVIDIA have announced a partnership that aims to significantly reduce both the costs and construction times associated with advanced nuclear reactors in the United States. This collaboration is part of the Department of Energy’s Genesis Mission, which is focused on accelerating the deployment of nuclear energy through innovative scientific computing.

As part of this initiative, known as Prometheus, the partnership seeks to cut reactor development timelines by 50 percent and operational costs by more than 50 percent. The goal is to leverage artificial intelligence (AI) alongside supercomputing to enhance the design, licensing, manufacturing, and operational processes for nuclear reactors.

Transforming Nuclear Energy Deployment

The Prometheus initiative aims to address the pressing demands for nuclear energy in a rapidly changing energy landscape, particularly in light of increasing electricity consumption driven by AI infrastructure. The collaboration intends to create a synergistic relationship between AI-powered nuclear design and nuclear-powered data centers, effectively fostering a cycle of mutual support between the two technologies.

At the heart of this effort is the application of generative AI models, digital twins, and agent-based workflows. These technologies will simulate and validate reactor systems before any physical construction begins. By utilizing decades of legacy nuclear data and experimental reactor operations at INL, the digital twins will enable more efficient modeling and quicker decision-making processes.

John Wagner, director of INL, emphasized the transformative potential of this partnership, stating, “This partnership represents a transformative approach to one of our nation’s greatest challenges for deploying abundant, reliable nuclear energy at the speed and scale required for our AI-driven future.”

NVIDIA will provide its advanced AI infrastructure and GPU-accelerated computing platforms to support the initiative. The collaboration also includes the acceleration of nuclear simulation codes, such as MOOSE, BISON, Griffin, and Pronghorn, on NVIDIA GPU architectures, which will help expand modeling capabilities and reduce computation times.

Utilizing Supercomputers for Enhanced Capabilities

The success of Prometheus will rely heavily on the Department of Energy’s leadership-class supercomputers, which will facilitate large-scale AI model training and reactor simulations. INL’s facilities, including the Neutron Radiography Reactor and the Microreactor Applications Research Validation and Evaluation project, will provide essential real-world validation data for the digital twin models.

The initiative aims to support regulatory modernization and promote the adoption of AI-driven nuclear tools across the industry. Future expansions may involve reactor developers, utilities, investors, and additional national laboratories.

Rian Bahran, Deputy Assistant Secretary of Energy for Nuclear Reactors, remarked on the broader implications of this partnership. “This is the moment to decisively advance AI-accelerated nuclear energy deployment, increasing America’s energy affordability while also catalyzing the development of artificial intelligence in the United States,” he stated. Bahran highlighted that this public-private partnership represents a targeted approach to AI acceleration, which has the potential to fundamentally transform how nuclear energy is deployed and how research and discovery are advanced.

This ambitious collaboration between INL and NVIDIA reflects a significant step towards making nuclear energy a more efficient and cost-effective option in the global energy landscape. With the increasing urgency to address energy demands and climate challenges, the integration of AI and advanced technologies in nuclear energy stands to play a crucial role in shaping the future of energy production.