Nuclear power is increasingly viewed as a potential solution to the surging energy demands driven by advances in artificial intelligence (AI). Yet, a novel approach suggests that AI might also facilitate the development and deployment of nuclear energy itself. Zavier Ndum Ndum, a Ph.D. student specializing in nuclear engineering, is at the forefront of this intersection, utilizing large-language models (LLMs)—the same AI technology behind chatbots like ChatGPT—to enhance nuclear engineering research and applications.
Ndum is focused on building a suite of tools that integrate the rapid knowledge acquisition capabilities of AI with simulation functionalities, one of which is a framework known as RADIANT-LLM. This tool aims to streamline the process of gathering information from technical databases and research papers, significantly reducing the time and effort required to search through extensive archives. Unlike generic chatbots, RADIANT-LLM pulls information from documents stored locally, ensuring that sensitive data remains secure.
RADIANT-LLM offers a user-friendly solution to the often labor-intensive tasks faced by nuclear engineers. By leveraging a technique called LLM augmentation, Ndum’s framework goes beyond simple query responses found in commercial chatbots. It efficiently sorts through both current and outdated documents to provide the most relevant and up-to-date information, which is particularly crucial in the nuclear sector where accuracy is paramount. “ChatGPT is incredible, but I would say it’s a jack-of-all-trades,” Ndum noted, emphasizing that while such tools can offer general knowledge, they may lack the depth required for technical inquiries.
One of the significant advantages of RADIANT-LLM is its transparency; the output generated includes citations, allowing users to verify the information against the original source. This capability is critical in an industry where regulatory compliance and precise details are essential. “The tools are not meant to replace researchers or engineers,” Ndum clarified. “They are designed to be expert-level assistants that dramatically reduce the time spent on tedious tasks.”
Ndum’s mentors at the College of Engineering see considerable promise in RADIANT-LLM, particularly in how it can expedite the licensing process with the Nuclear Regulatory Commission (NRC). Dr. Yang Liu, a nuclear engineering professor and Ndum’s advisor, highlighted that it can efficiently sift through extensive NRC resources, thereby alleviating the workload for human researchers. “We always want a human in the loop,” Liu remarked, underscoring the importance of human oversight despite technological advancements.
RADIANT-LLM builds on Ndum’s previous work with pre-trained LLMs, including the AutoFLUKA framework, which automates simulations by interfacing with the FLUKA software for radiation transport, and AutoSAM, which automates thermohydraulic simulations. His newest framework, AROMA-GPT, focuses on safely supervising advanced nuclear reactors. Ndum envisions a unified system that integrates these models to enhance simulation capabilities through AI agents.
These AI-driven tools are not only designed to assist researchers but also to improve the learning curve for new users. According to Liu, undergraduate students have found Ndum’s frameworks to be particularly helpful in quickly mastering complex simulation programs. “It has a reduced learning curve for new users, and improved productivity for experienced users,” he said. This adaptability underscores the framework’s potential as a transformative tool in nuclear engineering education and research.
Ndum presented RADIANT-LLM at the 2025 Institute of Nuclear Materials Management Annual Meeting, where his paper won the J.D. Williams Best Student Paper award in the Nuclear Security Division. “It’s a testament to the hard work that I’ve been doing, as well as the guidance that I’ve received from my mentors,” he remarked, reflecting on the recognition of his work. “It also shows the potential that generative artificial intelligence has in the nuclear domain.”
Looking ahead, the integration of AI in nuclear energy applications could redefine both the efficiency and safety of nuclear engineering practices. As technologies like RADIANT-LLM continue to evolve, they promise to enhance the capabilities of nuclear engineers, potentially reshaping the future of energy production.
See also
AI Study Reveals Generated Faces Indistinguishable from Real Photos, Erodes Trust in Visual Media
Gen AI Revolutionizes Market Research, Transforming $140B Industry Dynamics
Researchers Unlock Light-Based AI Operations for Significant Energy Efficiency Gains
Tempus AI Reports $334M Earnings Surge, Unveils Lymphoma Research Partnership
Iaroslav Argunov Reveals Big Data Methodology Boosting Construction Profits by Billions
