Federal Project Grant Award Summary

The National Science Foundation (NSF) Division of Computing and Communication Foundations awarded a $349,875 CAREER grant to the University of California, Santa Cruz, effective October 1, 2026, through September 30, 2031. This project develops a safety-aware learning framework to identify and mitigate risks arising from human interactions with large language models (LLMs) in healthcare applications. The research delivers three integrated technical components: (1) predictive models using supervised and contrastive learning to detect interaction risks including overreliance, stereotyping, manipulation, and privacy violations; (2) robust learning methods incorporating user intent, clinical context, and interaction dynamics through adversarial training and personalized reinforcement learning algorithms to mitigate identified risks; and (3) interpretable outputs designed to enhance the reliability and trustworthiness of healthcare LLM systems in sensitive domains such as mental health and addiction recovery.

In addition to research outputs, this CAREER award funds educational activities that develop interdisciplinary coursework and engage students from diverse backgrounds in research at the intersection of artificial intelligence and health. The work contributes to the broader NSF objective of advancing investigator-initiated research in computing and information science while supporting responsible artificial intelligence development and safer patient outcomes in healthcare technology applications (CFDA 47.070, Computer and Information Science and Engineering).