Text Retrieval and Generation for NRC Staff Support
Description
NRC staff are required to discover, retrieve, and analyze information from internally-managed and externally-provided documents. These documents can be lengthy as well as dense with content and often require cross-reference against multiple policies, standards, guidelines, and procedures. The process of discovery, retrieval, analysis, and cross-reference is time consuming and often represents the most significant bottleneck in carrying out agency activities and in making agency decisions.
Detailed example
The output is semi-structured content (e.g., text and tables) responsive to the NRC staff's queries.
AI / analytics pattern
Generative AI: AI that generates new or synthetic content (e.g., images, videos, audio, text, code).
Automation level / stage
c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency.
Expected benefit
This use case enhances the NRC’s efficiency and consistency in licensing, oversight, and other regulatory activities. It enables more efficient capture and use of institutional knowledge to mitigate against the risks associated with staff attrition. The NRC expects this use case to improve the timeliness of its risk- and policy-informed decisions, supporting its mission to enable the safe and secure use and deployment of civilian nuclear energy technologies and radioactive materials.
Controls / human review
ATO: Yes; PIA: Not published
Data needed
Where appropriate, prior NRC records are used in concert with subject matter expert review to evaluate the performance of the model's outputs.