The public comment period for the publication has closed. We are currently reviewing the comments received.
NCCoE Chatbot
An examination of opportunities to leverage generative AI capabilities to support the work of the NCCoE and share insights on the development process for organizations interested in the technical decisions, observed limitations, and risk-informed safeguards needed to adopt AI tools securely.
Project Abstract
The NCCoE is currently working to implement two potential applications of generative AI capabilities.
The first capability is the development of an internal chatbot to assist NCCoE staff with discovering and summarizing cybersecurity guidelines tailored to specific audiences or use cases. The NCCoE chatbot was built using retrieval-augmented generation (RAG)-based LLM technology. This approach combines techniques from information retrieval and natural language generation, enabling the chatbot to provide more focused, contextually relevant responses by leveraging a repository of cybersecurity knowledge. Specifically, the chatbot is currently designed to search NIST publications exclusively, enabling users to receive information that is aligned with the NCCoE’s guidelines and best practice documents.
The second capability is the development of a Profile Builder to provide an initial set of content for Community Profiles intended to tailor the NIST Cybersecurity Framework (CSF) 2.0 for specific communities or use cases. The Profile Builder is intended to be an internal application to provide a structured approach for defining the profile.
To further enhance its mission, the NCCoE identified a potential application for a secure, internal-use chatbot. The development of the NCCoE chatbot aims to assist users in discovering and summarizing cybersecurity guidelines specific to their needs.