Exploring the Capabilities of the Frontier Large Language Models for Nuclear Energy Research

Ahmed Almeldein, Mohammed Alnaggar, Rick Archibald, Tom Beck, Arpan Biswas, Rike Bostelmann, Wes Brewer, Chris Bryan, Christopher Calle, Cihangir Celik, Rajni Chahal, Jong Youl Choi, Arindam Chowdhury, Mark R. Cianciosa, Franklin Curtis, Gregory G. Davidson, Sebastian De Pascuale, Lisa Fassino, Ana Gainaru, Yashika Ghai, Luke Gibson, Qian Gong, Christopher Greulich, Scott Greenwood, Cory D. Hauck, Ehab Hassan, Rinkle Juneja, Soyoung Kang, Scott Klasky, Atul Kumar, Vineet Kumar, M. Paul Laiu, Calvin Lear, Yan-Ru Lin, Jono McConnell, Furkan Oz, Rishi Pillai, Anant Raj, Pradeep Ramuhalli, Marie Romedenne, Samantha Sabatino, José Salcedo-Pérez, Nathan D. See, Arpan Sircar, Punam Thankur, Timothy Reed Younkin, Xiao-Ying Yu, Prashant Jain, Tom Evans, Prasanna BalaprakasharXiv 2025 · DOI · arXiv

Abstract

Nuclear energy research is characterized by extreme complexity, requiring decades to transition from theoretical concepts to operational reactors. This study evaluates the potential of frontier Large Language Models (LLMs)—including GPT-4.5, Claude 3.5, and Gemini 2.0—to accelerate this process. Through a workshop at Oak Ridge National Laboratory (ORNL), 14 interdisciplinary teams applied these models to diverse fusion and fission challenges. Results indicate that while LLMs cannot yet replace domain expertise, they serve as powerful "scientific accelerants" for cross-disciplinary synthesis, workflow automation, and rapid prototyping.

Introduction

The development of nuclear energy systems involves "multiphysics" expertise, necessitating the simultaneous management of plasma physics, neutronics, materials science, and thermal hydraulics. The inherent complexity often leads to protracted development timelines. This research investigates whether the latest generation of reasoning-capable LLMs can compress these timelines by identifying research gaps, generating simulation code, and designing experimental programs for next-generation reactors.

Methodology

The study utilized a "human-in-the-loop" approach across 14 technical challenges. Teams leveraged frontier models such as ChatGPT-4.5, o1, o3, and Claude 3.5 Sonnet. The methodology focused on structured workflows rather than simple prompting, employing iterative refinement and Retrieval-Augmented Generation (RAG) to interface with scientific literature and simulation frameworks like XGC, NekRS, and OpenMC.

The structured AI-driven workflow for nuclear energy research, from high-fidelity simulations to model fine-tuning.

Key Insights and Technical Synthesis

The Knowledge Bridge

LLMs demonstrated a unique capability for cross-disciplinary synthesis, connecting disparate fields such as plasma control and materials science. This "knowledge bridge" allows researchers to identify solutions that might otherwise be overlooked in siloed environments. For example, models successfully connected plasma control problems with materials science solutions during the intensive workshop sessions.

Workflow Automation and Rapid Prototyping

A primary strength of current LLMs lies in designing experimental programs and workflows. For instance, the "Foundation Model for Fusion" team produced a technically sound concept paper and LaTeX-formatted workflow within hours. Similarly, AI-directed experimental designs for Molten Salt Reactors included proposed budgets and timelines that would typically take weeks to compile manually.

Code and Logic Generation

While the models struggled with production-ready code for highly complex solvers, they were effective at generating Mathematica scripts and structured outlines for simulation input decks. In material degradation studies, AI-generated models achieved a predictive accuracy of $R^2 \ge 0.90$ using acoustic and strain data.

Results and Discussion

The workshop outcomes across the 14 domains are summarized in the following table, detailing the specific challenges, outcomes, and limitations encountered.

Summary Table of the 14 nuclear energy research challenges addressed during the workshop, highlighting key outcomes and limitations.

Limitations: The "Hallucination" and Data Barriers

Despite their strengths, LLMs are prone to physical hallucinations, such as suggesting impossible cooling fluid velocities (e.g., 50,000 m/s). Furthermore, performance is significantly hindered by "data silos"—specifically paywalled literature and export-controlled software (e.g., SCALE or MCNP) that the models cannot access. There is also a notable lack of source transparency in AI-generated images, making it difficult to verify material-specific defects.

Conclusion and Future Directions

The current generation of LLMs acts as a "complement" to expert validation rather than a replacement. Future research should prioritize the development of "Nuclear-Specific LLMs" trained on curated, secure datasets. The ultimate goal is the creation of "Agentic Workflows" where autonomous AI networks coordinate theory, simulation, and experimental validation in a continuous feedback loop to drastically reduce the time-to-market for clean nuclear energy.

Summary updated

BibTeX
@misc{almeldein2025exploring,
  author = {Ahmed Almeldein and Mohammed Alnaggar and Rick Archibald and Tom Beck and Arpan Biswas and Rike Bostelmann and Wes Brewer and Chris Bryan and Christopher Calle and Cihangir Celik and Rajni Chahal and Jong Youl Choi and Arindam Chowdhury and Mark R. Cianciosa and Franklin Curtis and Gregory G. Davidson and Sebastian De Pascuale and Lisa Fassino and Ana Gainaru and Yashika Ghai and Luke Gibson and Qian Gong and Christopher Greulich and Scott Greenwood and Cory D. Hauck and Ehab Hassan and Rinkle Juneja and Soyoung Kang and Scott Klasky and Atul Kumar and Vineet Kumar and M. Paul Laiu and Calvin Lear and Yan-Ru Lin and Jono McConnell and Furkan Oz and Rishi Pillai and Anant Raj and Pradeep Ramuhalli and Marie Romedenne and Samantha Sabatino and José Salcedo-Pérez and Nathan D. See and Arpan Sircar and Punam Thankur and Timothy Reed Younkin and Xiao-Ying Yu and Prashant Jain and Tom Evans and Prasanna Balaprakash},
  title = {{Exploring the Capabilities of the Frontier Large Language Models for Nuclear Energy Research}},
  year = {2025},
  eprint = {2506.19863},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2506.19863},
}
Figure