Frequently Asked Questions About AI
On this page:
Index to All Frequently Asked Questions Pages
How is the NRC preparing for the use of AI in NRC-regulated activities?
Significant industry interest in deploying AI technologies means a need for regulators to be prepared. Regulatory preparedness is a critical part of the NRC’s plan to support AI deployment. Nevertheless, the concept of being ready to regulate something "new" is in itself "not new." The NRC anticipates that the U.S. nuclear industry will deploy AI technologies that may require regulatory review and approval in the next few years. In May 2023, the staff issued "Artificial Intelligence Strategic Plan: Fiscal Years 2023-2027," NUREG-2261, to ensure our agency remains adept at effectively evaluating and reviewing future AI applications. This plan also serves as a platform for engaging in conversations centered around identifying and resolving regulatory issues for AI deployment.
What regulatory actions are you considering involving AI?
The NRC is following the rapidly evolving landscape of AI. The staff developed the NRC’s AI Strategic Plan (NUREG-2261) to ensure readiness for regulatory decision-making. The plan includes the goals of establishing an organizational framework for AI review, strengthening AI partnerships, cultivating an AI-proficient workforce, and pursuing use cases to build a robust AI foundation. Through these efforts, the NRC is committed to ensuring the safe use of AI technologies while maintaining public trust and confidence in nuclear safety. As part of its public outreach, the NRC hosted an AI public workshop at NRC Headquarters on September 17, 2024, to discuss potential regulatory gaps and considerations in evaluating AI technologies.
How is the NRC engaging with external stakeholders with respect to AI?
The NRC hosts an annual Data Science and AI Regulatory Applications public workshop to provide a forum for the NRC, nuclear industry, and stakeholders to discuss the state of knowledge and research activities related to data science and AI and their application in the nuclear industry. At these workshops, the NRC works with internal and external stakeholders to identify the benefits and risks associated with the use of AI in regulatory activities and discusses ongoing and planned projects in the nuclear industry. Based on feedback from these workshops, the nuclear industry could start deploying AI in nuclear applications in the near-term.
How is the NRC engaging internationally on AI topics with regulatory counterparts and stakeholders?
Staff is actively involved in international activities and maintains a strong leadership role in nuclear safety research on AI while gaining insights into international experience and best practices. The NRC is involved in an ongoing collaboration with the Canadian Nuclear Safety Commission and the United Kingdom’s Office for Nuclear Regulation on an AI regulatory principles paper and potential areas of research. The NRC published the paper on September 5, 2024, and plans to share it with the International Atomic Energy Agency (IAEA) and other interested parties to further facilitate dialogue on AI. The NRC is also involved in two ongoing working groups with the IAEA, focusing on the safety implications of using AI in nuclear power plants and deploying AI solutions for the nuclear power industry. Emphasizing our commitment to these activities, the NRC hosted an IAEA technical meeting on deploying AI solutions for the nuclear power industry at NRC Headquarters in March 2024, following the NRC's 2024 Regulatory Information Conference.
How is the NRC advancing the use of AI for internal activities?
AI presents an opportunity for the NRC to improve operations and support for its mission. Recently, the NRC published SECY-24-0035, summarizing potential AI applications and the staff’s overall approach to effectively leverage AI at the NRC.
The NRC identified 61 potential use cases across the agency that have the potential to drive value and enhance how it meets its mission. From these 61 potential use cases, the NRC identified 36 use cases that align with the capabilities of current AI tools, while the remaining 25 could be addressed using non-AI solutions. Among the 36 recommended AI use cases, 16 had an overarching theme of leveraging generative AI, including commercially available productivity tool assistants. The other 25 use cases focus on aspects of data analytics or process automation that could be achieved without the use of AI.
To effectively use AI and ensure long-term success, the NRC will develop a framework to integrate AI into the agency's operations. This involves two strategic next steps: (1) developing an enterprisewide strategy to advance the use of AI within the agency and (2) investing in foundational tools to support the implementation of several use cases. As the agency begins implementing the use cases, the NRC will prepare and share an inventory of AI use cases with the public and other Government agencies.
How is the NRC enabling improved tools such as the Agencywide Documents Access and Management System (ADAMS) to enable development of AI solutions?
AI played a role in transforming our internal database known as ADAMS. The NRC used AI-powered cognitive search technology to create an enhanced ADAMS search engine referred to as ADAMS Content Search which significantly streamlined search processes for the staff and provided a more modern search interface similar to commonly available public web search capabilities. This project offered valuable insights into practical uses of available AI tools. The NRC plans to launch a public version of this tool, known as ADAMS Public Search, in late 2024.