This $600,000 federal Project Grant award was provided by the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070). The grant will support research at the University of California, San Diego (UCSD) to explore limitations and develop new approaches for using conversational language models to make personalized recommendations across various applications, such as movie suggestions, personalized clinical support, and other settings. Key focus areas include collecting new datasets, ensuring fairness and bias mitigation in language model recommendations, and developing simulation approaches for offline model evaluation. The project aims to advance research in areas like recommender systems, natural language processing, and AI controllability, ultimately enabling new applications that combine language and user interaction data. The grant period runs from April 1, 2025 to March 31, 2028.