AI to Streamline Nuclear Reactor Licensing Process
The Department of Energy and Microsoft are collaborating to utilize artificial intelligence to significantly reduce the administrative burden and time associated with licensing and permitting new nuclear reactors.
The initiative, stemming from a discussion at the Idaho National Laboratory, aims to leverage AI to automate the generation of engineering and safety analysis reports – standard requirements for applications to the Nuclear Regulatory Commission (NRC) and the DOE. Currently, compiling these reports is a time-consuming and costly process for reactor developers. “You theoretically, too, could have some of these models check your work, make sure you didn’t miss anything that’s a requirement and things like that,” said Chris Ritter, division director for scientific computing and AI at the Idaho National Lab. The effort doesn’t involve the AI performing engineering analysis, but rather organizing and compiling existing data for human verification.
According to a previous paper by the Idaho National Lab, AI tools could potentially improve processing time by 21%, but recent advancements in large language models like those from Anthropic and OpenAI suggest improvements could reach 30% to 50%. This comes as the Trump administration’s support for nuclear power has spurred innovation in related processes. The Energy Department released a report in September detailing the need for expanded nuclear energy to meet national electricity demands, identifying licensing documentation as a key obstacle. Streamlining this process is crucial as the NRC aims to reduce review times for new designs to 18 months.
This project builds on previous work by the INL and Microsoft, including the development of the world’s first nuclear reactor digital twin in 2023, a virtual replica of a reactor at Idaho State University. Officials stated the next step is to verify the AI tool’s performance in a government cloud environment and quantify the efficiency gains.
Researchers will continue to test and refine the AI tool, with the goal of improving efficiency and enabling more dynamic solutions for the nuclear energy industry.
