Online experiences enrich our lives in many ways from seamless shopping to staying connected with friends and family on social media. Yet, online experiences also can be challenging. For example, online health advice can be confusing and unreliable. Malicious apps, including health apps, are a safety threat because they can steal confidential information. The recently introduced large language models may amplify both these experiences. The goal of this project is to combine rigorous engineering and computer science techniques to devise methods to enhance the safety and experience of online activities. This project will provide an interdisciplinary environment to study various safety issues related to online activities. Data including physiological data from subjects, and online activity will be acquired and then analyzed using machine learning and other techniques for modeling online activities and their impact on the users. Physiological data from wearable devices, virtual reality headsets, non-invasive brain imaging techniques such as electroencephalography and Functional Near-Infrared Spectroscopy will be collected while individuals are engaging in online activities. Machine learning tools, including large language models and federated learning, along with other techniques such as stochastic state space models will be used for data analysis and system modeling. This site is supported by the Department of Defense in partnership with the NSF REU program. This awa