This project will explore the potential of AI and LLMs to be leveraged in the effort to gain improvements in both quality of insights and efficiency in large-scale efforts to collect, process, understand and respond to feedback and comments from constituents. The initial effort will focus on the NJPACT REAL rule proposal that is expected to generate multiple thousands of comments. One anticipated deliverable is a general tool into which any large file(s) of comment text may be loaded, that can effectively answer a wide range of questions about that set of comments. Other tools to be considered include those that can streamline the production of formal responses for publication and role-playing tools to anticipate comments ahead of proposals.